Techstrong TV November 17, 2025

Hey, everyone. Remember when tech used to fight the good fight? They were the good guys.

You're watching Techron Gang. Hi everyone. Happy Monday.

It's Alan Shimel. Guys, I am glad to be home if only for a day. Uh, it was, it was cold in Atlanta and I, you know, it was also really bad wifi at that conference.

It was, it was hard. It was hard maneuvering and working, but it was a great c**n. I've written about it.

We spoke about it all last week on our gangs. You could check it out. Um, but we're back here in studio and we've got a great lineup of our gang members to talk about some good stuff.

Let me introduce you. We have a, a DevOps dozen finalist, Garima Boal, a security analyst, Jack Poller, an analyst who's finding her next career, or her next joy in life, our friend Kimberly Bates. Would that make you like a recovering analyst, Kimberly?

Yeah, yeah, yeah. A recovering analyst. There's hope.

That's good to know. Yeah. And there's hope.

And then, um, and then two more analysts. Were always with us, the ones and only Mitch Ashley and Guy Coer. Guys, how are you?

Good, good. So, I don't know, maybe it was the kumbaya of the open source movement and the idea of doing things for the common good that got me, or maybe I'm just tired of seeing the pig's feet at the T trial, but I, I, I did a, a shimmy says, uh, an I it article and a LinkedIn article last week on this idea of, you know, tech used to be the good guys, even in the movies, right? You go to the Avengers, I'm Iron Man, right?

He was a good guy. He was fighting crime, fighting the good fight. He, to me, that's what Elon Musk should be aspiring to be, right?

Or, or any of these tech mogul, tech bro billionaires that we have kind of in, in place today. But that's not the tech we get today, right? Increasingly, you know, it's a, it's a selfish kind of egotistical egomaniac or egomaniacal, if you will, kind of driven industry.

We don't have any numbers because the government's not releasing any numbers, but stories, you know, garnered for from like employer records. And, you know, some of the big PEOs and stuff say that we might have lost maybe 35,000 more tech jobs last month. Last month alone.

Things aren't so great in tech, in spite of the fact that we're spending record money building data centers in this AI thing. And I, you know, increasingly people are recognizing, and we'll talk about it later, you know, we might be in an AI bubble, who knows, but Mitch, guy, Kimberly, Jack Garima, we've all been in tech collectively. There's a hundred years, maybe 200 years worth of tech experience here.

Have we, have we, are we not the good guys anymore, or were we ever, maybe we were never the good guys, But first came to mind. Um, you know, we've all had people that we've admired over the years, whether you're like Steve Jobs or whoever it might be, right? Alan Kay is a researcher.

Um, I think we're, we're in the Gordon Gecko phase of tech leaders. Like, where's my fricking trillion dollars? Greed Is good.

Yeah, exactly. For my bonus. And, you know, I, I think for a while, Elon Musk was held up as, you know, someone that's admirable because of cool things that he was doing with space and, and, uh, you know, Tesla, things like that.

Now he's just kind of another one of the greedy people on, on the block and, uh, doing, doing side deals with everybody and, you know, forcing what he wants out of his board, just to pick on one. But I, I don't know if we were ever really, truly, you know, tech were the good guys necessarily. I'm not saying we're the bad guys, but I think we had people that we admired and we held up as whether role models or not, I don't know, but people that we just admired a great deal because of what, what they did for the industry and their leadership in the industry.

That's what I think we're missing. So I'm, I'm gonna be the contrary here a little bit, because the first thing that came to mind is that when I joined IBM way back when, and I represented at least a quarter of the years that you talked about here, No, you don't. We're all your age stuff close To it.

Close To a quarter. The three of us make up three. We Were on our antitrust investigation and we're about to get slid up into three.

And there was a reason why they were doing that is because IBM was not playing nice. We were wrapping up the things, the big, you know, the burrows, the Honeywells, that kind of stuff. We were screaming over them.

I mean, and then I started thinking about, you know, Tom Watson, he had his entire desk dropped in the front yard, uh, NCR when he worked for them. That was not exactly very mice either. And then you add the, okay, okay, 1998, I was looking this up, Microsoft versus a Mosaic.

Netscape, you had 20 states going after Bill Gates. Bill Gates is saying crush him. Just crush him.

You know, he was really nice back then. Now he's like a really good, nice little woke guy and was like, well, baloney. Sorry, I almost went the other way.

And then, then I'm gonna add one more to my favorites, is Uncle Larry take no prisoner, Mr. Larry Ellison kind of guy. I mean, he's been dropping off people every quarter.

Think It's Grandpa Larry, by now, He's grandpa, sorry, grandpa, Larry, whatever. No, that it's media mogul, Larry, Media mogul, Larry, Larry, whatever. I mean, every quarter you wipe, he wipes out a, you know, a percentage of his people there.

And that's been going on forever. And he, he just never has to publish it because he does it every quarter the same way that, you know, uh, the guy at GE did it. You know, it's the same kind of principle.

So, and like, you know, I think what happened is that we went, and I'll shut up after this. We went through this period where everybody was like, let's be nice. Let's be sweet, let's be okay.

And now we're back into let's go compete. And, um, that's, that's where we're at. And you know, I embrace it.

I embrace the competition. And, um, that's, Yeah. I, I don't, you know, I, I've only, let's let's let everyone else talk before I jump back in, but thoughts from the rest of the team?

Well, I'll, I'll, I, I, I will say that I'm on Kimberly's side and, and having been both at IBM around the same times, and then at, uh, Netscape during that time and, uh, new employee orientation at Netscape was a speech by, uh, our now venerable vc, mark Andreessen, who said, uh, Netscape earns $70 million a year because from ads sales on our homepage, because nobody ever changes the homepage. And so we're suing Microsoft over the browser to protect that revenue. That was all it was about, right?

Is protecting that revenue. It had nothing else to do with anything else. And they knew they sued Microsoft just because they could.

And when we, I did an IETF meeting where we were discussing, uh, um, uh, TLS, the TLS specs, and we had two coalitions, the pro Microsoft Coalition and the anti Microsoft Coalition. And we spent most of the time figuring out what Microsoft wanted and trying to design the protocol to make it more difficult for Microsoft. Well, when was done low level Engineering level, I'm sorry, windows ain't done till, notes don't run, right?

That was the old thing at Microsoft back then with notes, because the 1, 2, 3 was killing Excel. But, but guys, you, you're missing, I, I think you're missing my point with all due respect, you know, to quote the infamous Don Barini from Godfather, after all, we are not communists, right? There's nothing a matter with making a profit.

There's nothing a matter with competing. Good old American competition may the best person and company and product win. There's nothing a matter with using the law to your advantage.

This is how the rules of the game or played. This is, and there's, and that's great. That's America, that's capitalism.

That's market forces at work. Don't fool yourselves into thinking that's what's going on here. What we have here is what we call collusion.

What we have here is Uncle Larry is in bed with Young Sam and Young Sam, as soon as he gets outta bed with Uncle Larry runs over to, to, whether it's the Microsoft or Google guy, and does his little jig there, and they pass money around this virtuous circle of the eight or nine of them creating barriers to entry for the two guys for the next was, and jobs sitting in a garage working on something or, or, or anyone else who wants to come in here. Right? So how is that so much different than what was going on in Sand Hill with the VCs?

I mean, that the VCs are just rolling doors of your, you know, it's, it kind of goes back to New York when you had the Morgans Morgans and the, you all the, the people back in whatever years it was. And those were the people that socialize with each other. And that's the same way it is right now.

There, There The detectives, But not at this level. And then I'll tell you what else is the big difference. They've got a new partner in the circle who's taken a piece of the action, right?

The federal government, the federal government used to, and the court system used to be the, the arbitrary, the, the, the, you know, the guys who made sure at least, even if it wasn't real, at least the facade was Horatio Alga work hard. You'll get ahead now. They got their thumb on the scale too.

I, I feel like the, the, the discussion here has shifted because my take on your take, Alan, My take on your take and Your take. Go ahead. Yeah.

Well, and, and how you summarize it here and, and your shimmy says, and how you summarize it here was you were talking about, um, tech to benefit the world. Tech used to be the good guys to benefit the world. Now it's the bad guys to benefit these nine people.

That's what I thought you were talking About. Yeah, no, and, and, and you are right, guy. We, I I, shame on me for not mentioning that too.

I think there was sort of this altruistic of we're changing the world with tech, right? Tech is the great kind of equalizer. We're going to get women and other underrepresented in groups, you know, working in it.

And, you know, and, and Kimberly look, I mean, no, I'm not saying anything bad or I don't hold it against me, but you came up in a time when it was really hard for a woman, too. No, you don't think it was hard for you. I, I am, I'm one of those contrary people on it.

I, I didn't hit the walls like I felt like, you know, other people talk about just either that or I just ignored it. It was Completely, well, God bless you and good for you, Clueless. I was probably just clueless to it.

It just kind of blinked, Oblivious, dude, that all, they just do your job. Karima, what about you? Do you feel like, Yeah, I think, uh, you made the point very clear in the beginning that, you know, tech for good, tech for communities, and I represent community and community leadership here.

So I would say that, you know, with all this, what you wrote, what are the repercussions happening? It's fear, it's disengagement, it's normalizing corruption, uh, stifled creativity. And what happens when it goes to the peak communities, uh, you know, strike back and this is what will happen.

And I will give you an optimistic view on this, because I think you mentioned about open source movement. I come from that open source. I am, um, I'm the child for, from the open source movement, the communities.

And I feel that, you know, um, it's time for us to kind of ensure that something what happened in the 1980s with, you know, uh, Linux, uh, taking the power back, right? With community and open source, uh, trying to kind of, uh, uh, steer the needle in the right direction. I, I, I am more optimistic with leadership and the communities and history reminds us what happens, happened with DevOps, you know, all those tools, capabilities, this, uh, Kubernetes movement.

How did it got started? You know, it's massive right now. So I'm more optimistic.

It's just a matter of time. I mean, you cannot stifle innovation and adaptability on the name of egg diplomacy. And this, I mean, it's just, you know, how we see it today and what can happen when community comes back and try, tries, tries to take charge of it.

Amen. I I think I'm quite optimistic about this development that I think you pretty accurately described, Alan. Um, I kind of agree with everybody here, which is really weird.

Um, the, so I think the fundamental issue here, I think is that there's tech and there's science, and they are, they are very well linked together. The science brings the discoveries, like the original invention of the semiconductor that enables the tech, namely, you know, information technology scientists are pursuing knowledge, truth, betterment as a general rule. Yes, they have, you know, businesses as well, and they, they, they do commercial things.

But that's the general culture there. Um, we're, we're gonna see that this week. I'm attending Super Compute, which is one of the great fusions in the world of science and technology.

Technology or tech. That's a business like you described. And I think that we are reverting to the mean of understanding that the business is in the business of its own business and profit and science remains the pursuit of knowledge fundamentally.

And all that happened was probably led by jobs, was this idea. I mean, Steve Jobs was this idea that, that, um, there's an altruism. I mean, that's one of the things I despise about the whole Silicon Valley culture is how this new platform, uh, and and Gizmo that I've invented is what's gonna bring peace to the Middle East and extend lifespans by 10 years and all this other kind of garbage that helps the VCs get all excited and the money flow, but it's really just garbage.

And so if we're gonna finally start recognizing them for what they're trying to do, which is to make themselves rich and get their trillion, then, you know, great. 'cause that's what they've been doing all along. They've just been papering it over with this idea that tech is also in the business of the common good, and it's the science that's in the business of the common good more than the tech.

But, but they, they, but as you start it off with, they're intrinsically linked, right? That's the tension. That's the tension.

So it's always been there, but you know, it's kind of like adopting this idea that this innovation, which is the word usually used, this new widget is, uh, like the, a scientific pursuit for the good of all. And it's not, it's not, it's the new widget that benefits from the science. We're seeing that in AI right now.

You, You've hit it right on, which is that we in the Silicon Valley and the Silicon Valley tech leaders have put a fig leaf of altruistic respectability on top of what they do, because you, and there's a lot of reasons for that. But it's, this is history repeating itself over and over and over again. If you just look at the printing press, when the printing press was invented, it was a, I don't know if it was created for altruistic ideals, but it had a very, um, you know, it ushered in the age of enlightenment where, you know, in the education and the explosion of knowledge throughout the world.

But then if you look many years later, we had a closing down of information availability and our knowledge and media was controlled by three corporations in the United States for a very long time. And before that was controlled by William Randolph Hurst and the newspaper barons. And the explosion of the internet was wet, freed information out.

And now we can get news from a gazillion different channels. Look, Look how we are. Yeah.

But, but it's a cycle that repeats itself, right? So, you know, I I, I understand your frustration, Alan, but I think it's a little bit naive to think, oh my God, That this is different. Well, that's, maybe it's naive.

Let me ask you a question. Do you think Hurst was a good guy for what he did? None of these people are, are good guys.

I Think first points of bad they had none of Them. They had good No, but Our Features forever. I mean, what, what are your opinion of the, let's beside the point, let me just tell you, hear me out their name by itself.

Don't let it prejudice your thought, but the robber barons, they're called Rob, but do you think was a good guy? Do you think Getty was a good guy? How about Mellon?

Yeah, those Melon was one of the robber barons. I Mean, we have a reason why we have antitrust laws. Exactly.

How about, it's the reason we used to have antitrust laws. But, but, but then, And then they make all this money, then they do this thing, this, you know, foundation and start giving away zillions of money Or, or build universities. Mylan la Mater was Carnegie Mell, excuse me, there you go.

Carnegie Mellon University, right? It was funded by the Robert Barons Or, or, or funded that building at the, at the university, university or whatever, put my name on it, right? It's like, Yeah, we've shifted from, we don't call him.

I've seen it's the same thing. Robert Baron. We call him Titans now, right?

No, I still call him Robert Barons. Mitch. No, we, yeah, some of, I mean, I, I mean Benioff, mark Benioff, who's the billionaire who, uh, does Salesforce, right?

Uh, he is single handedly funded a huge amount of the, the UCSF children's hospitals. And, um, uh, Ken Langone who did, uh, home Depot is the Langone Cancer Center in New York, which is very good at treating cancer patients. I mean, the, those billions of dollars that they've created have not gone entirely to gold toilets and lavish dinners.

And you know, Larry Ellison's yachts, yeah, he does have a yacht, but he also employed a couple thousand people, people That yacht after those, after the gold toilets and the lavish dinners and all that other sort of stuff. There's some money left over for cancer. That's great.

I'm not, I'm a capitalist. I'm not, I'm not trying to critique the entire system. I, I just, I'm more focused on this idea that tech and the tech industry does do good.

It should be, we, we benefit from it. But, um, expecting brands and companies, what, what I, what I like, what makes me optimistic is that I feel this cultural shift that you're detecting, Alan, where we stop believing when the next Google has their version of don't do evil, that we just stop believing that it's beside the point, it's marketing. And if the marketing stops working, we don't have to put up with it anymore.

Well, I think we could learn from mothers not in tech. Um, I'm gonna represent my home state. I admire Oracle, not Larry Ellison, Oracle, but the Oracle of Omaha.

So was he from western Nebraska? Western Omaha. Okay.

Yes. It was a little more Nebraska west of Western Oma. That was, that was a joke that we had.

No, we weren't talking about that in Q Con. Mitch says he's from western Nebraska. Warren, Warren Buffet.

You know, you talk about somebody who's made people a lot of money, including himself and and his investors. But you know, that's someone who, he, he had, he had a, uh, an attitude of not only philosophy investing, but about, because it's America, we are gonna do it this way, right? We're gonna invest in the things that, that Americans want are gonna need, whether it's the railroads here or candy that's sold in the, in the, in the airports here, or utilities, or, I mean, he was looking at, you know, it, it didn't matter if it was tech or not, he wasn't a big tech investor, but he was looking out as what really, what does the, the economy need?

Where is it heading? And how can you position yourself to be at that place when you know, kinda when the puck is, is sent that direction? And you don't have to be, you don't have to be the greedy bastard, I'll use the word word, who only cares about having the most money compared to all the other greedy bastards.

And that's kind of the, that's the, that's the environment. Whether you're talking about tech titans or you're talking about robber barons or what, that was all competition amongst each other of trying to either take the other person out or having the most stuff. Right?

You know, you don't have to do that. I mean, what he, he certainly look, he's on a pedestal above these other folks in terms of this. Here's my point on it.

It's okay. It's okay if that's what you want, want to be, right? But I, I think we all have to recognize that unfettered market economics, unfettered capitalism without some restraint is not good for the every man is not good for the gen pop, right?

However, neither is some sort of socialistic, communistic thing where, you know, we cut all the corn at the same height, so no one has more than anyone else. Our country has thrived historically as a mix, right? We, we do have market forces that shape our economy and shape our, our daily lives.

But the idea is that through lessons we've learned from the, the Gilded Age, again, they call the Robert Baron age, the Gilded Age, right? Through lessons we've learned through the depression from the Taft Hartley Act, and, and what that wrought through all of these, you know, historical things, we've come to say, Hey, we need a balance. There has to be a balance.

There has to be a, it can't just be, I buy my way in, I settle my a hundred million dollars lawsuit by giving you 25 million, and you wink wink, and let me do whatever I want it. It can't, in order for our system to work, right? We all, or most of us, have to believe that there's some sort of level, level pa playing field that we have a shot at the brass ring, that hard work ingenuity and smarts are rewarded.

It's not, it's not stacked up by some old boys game, you know, sitting back in the parlor. And all I'm saying is that we had it, it seemed like that's where we were for a while, right? Kimberly, to your point, IBM was on the verge of getting broken up.

Ma be did get broken up, right? Microsoft got hit hard, some might say, uh, for what, what they did with the browsers and windows and everything else. I don't think that's America today.

I, I don't think, you know, and because the other thing I think about it, and, and we will talk about it in, in another segment, is because we have made tech a strategic national imperative, and we've weaponized it. We're weaponizing it against our perceived enemies. But we can talk about that on another segment of the gang.

We gotta take a break. We're gonna come back and talk about AI anxiety on Wall Street. You're watching Textron Gang, You've earned it.

The spotlight, the responsibility, the weight of teams, companies, and entire industries fall on your shoulders, lives depend on your decisions. Your home life included that work. You are protected physically and digitally.

Nothing gets through your team without a fight. But in a globally connected world, everyone sees you, including those who mean to cause you and your organization harm. And now home your sanctuary attackers see an opportunity.

Your digital front door is wide open. And what compromises your home can breach your boardroom. Because the devil's greatest trick isn't targeting your workplace firewall.

It's convincing you that your personal life isn't at risk. Black cloak, digital executive protection, defending the new attack surface your personal life. Hey, everyone, we're back.

Wow, that was quite a discussion. Uh, we, it's almost a continuation of this though. But you know, maybe, maybe our friends down on Wall Street are feeling a little of this, you know, we used to fight the good fight too, and they're getting a little shaky and nervous.

Mitch, what do you think? Well, uh, you know, Everybody is talking about the AI bubble doesn't know whether, you know, what, anything about AI or not, right? It's common.

Like, are we on the cusp of having a big correction, right? Over 2%, uh, two to 3% correction. And everybody's anticipating that.

And I think almost that looking for it causes it to happen. I mean, we're recording this on Friday and had opened the, the markets down four to 500 points. Um, and there's lots of reason for it.

But, uh, n uh, Nvidia got hit yesterday, um, dropped in its value. And, you know, thing as if we keep on this kind rollercoaster, things will be pushed back up and then something will draw it down. I think it's, it, it's, we're at a place of saying, is AI real?

And how much is, have we overt, rotated if we've over rotated in investment in data centers and chips and stocks, um, in anticipation of ai? Or is it real? You know, is it, is it gonna deliver the value and it's gonna, it's those investments are worth, worth it?

And if history does repeat itself, you know, we've over, uh, invested in data centers before in the cloud, early, early in the cloud era, and we've overinvested in other things too. So it's not unheard of it. I think that concern is real, but what's, what's also bolstering it is the job layoffs that don't get reported by the government now, but, and especially in the tech industry.

And so I think people are concerned. It's, it's a time of uncertainty about where this is headed in our, as the financial stability of the country or ourselves, frankly, you could argue that's what some of the results of the election, midyear election showed, showed. So midterm election.

So Mitch, you, you know, you said, when that we sort of look for it and that's causing it, and I think that, you know, I looked, I, I did a Kimberly thing, and I actually looked this up, uh, uh, December 5th, 1996, Alan Greenspan gave a speech to the American Enterprise Institute and coined the term irrational exuberance. And I think what we're doing is not looking for it, but recognizing it now before it becomes really irrational and saying, you know, something doesn't smell right here. So, and I don't remember the exact numbers, but I believe I heard that, uh, open AI has made commitments for a trillion dollars in capital expenditure spend with something like, uh, $30 million in revenue, right?

Or 30 billion billion, 20 billion. They're expecting 20 billion hundred billion in revenue, right? So that's just, I mean, that's just such an jack.

I think It was only 800 billion, not a trillion. That's amongst friends. I mean, come on.

But it's, it's just so unbalanced. It's like these are unreal numbers that how does anybody make a commitment to spend that much when they have the capacity to pay it back, is just absolutely non-existent. Yeah.

And that's where's the irrational and supers we're recognizing. I, Jack, I just got a quibble with you saying that it's not irrational. Uh, uh, maybe I mean, too technical, irrational meaning divorce from rationality, um, the, what people are spending is based on what the market's at right now, but it's not a rational amount.

It's based on what they, it's based on their, their vibe that this is super important. And so in this capitalist, you know, system that we have, that we were talking about on the last segment, um, uh, people pay what they think they should pay, but they, they don't have a clear idea of what the return is going to be. And I, I don't have the numbers in front of me.

There have been recent studies, um, uh, or recent, what looks at this sort of thing from, from analysts for more qualified for that than, than I am. Um, say something like, you know, the best return we can expect is something like 20% on the, the, the level of investment in AI right now. And that's just a fact of the people paying the market rate for stuff without having a clear idea of what, what the return is gonna be.

So that's irrational. I think the irrational alluded No, I, I did, I I'm sorry I did real quickly, Mitch. I believe it's irrational.

I'm just saying we're recognizing it very early in the cycle rather than when everything has already collapsed because of it. Yeah, I think I, what I would take exception to what you're saying is the rational part of it is, is continuing to doing what we're doing, knowing that some of this is doesn't make sense. com go go to in an IPO in the ni late nineties for some ridiculous, absurd amount, right?

That's irrational, even though it happened, even though people made decisions that could, that, uh, helped that happen. And that's the question, are we in that place now? Is the overextension of commitments around?

Is the buying a buying, buying of business that's happening by co-investing across companies, you know, is that causing, you know, an, an overextension in, in the market and what actually what things are really worth? And the answer's probably yes. And there'll be a correction, and that's what happens.

And, you know, so, You know, the thing I've learned about bubbles over the years, I'm gonna, I'm gonna quote the great sage, Bobby Balala. Tony Soprano's, right hand man. Tony, what's the Godfather, man?

Okay, go ahead. Tony. Tony once says, Bobby, to do you think, do you think you, you feel it when, when they come and hit you?

When, when you get shot and Bobby says, you probably don't even see it coming. You probably don't even see it coming. Most bubbles.

Last week we were touting here how great this was, and the let the good times roll baby. And then before you know it, that's, that's how quick these bubbles burst. You don't even feel it.

You don't even see it coming. You don't feel it. And then it's, here is this, that moment I wish I knew I would be out with Warren Buffet somewhere, taking it easy, but I'm not, I'm here working.

Um, but, but I, I think what we're all saying is we do realize there's a little irrational exuberance here. There might be a payoff down the road, but the way our market, you know, back in the day, wall Street was ruled by institutional money. It was big, rich people or big companies, institutions that moved stock.

Today's Wall Street, 40% of your, of your equity business is retail. It's people like us, not that we're Warren Buffet or anything, but it's small time people like us. And, and generally, you know, the law of the herd plays into these things, and that, and that might be what we're seeing here, right?

Where, you know what they always say, follow the smart money. Who the hell, if I knew where the smart money was, I wouldn't be here still too. But, you know, that's what they say.

Follow the smart money. Um, Kimberly, I know you had, you had written some, some. No, no.

I only comment is I'm welcoming a correction right now. I I think it's, to my personal opinion is not that I'm a super, I'm not a really good investor or whatever. I, I hand it over to some other people to take care of that stuff, but I believe that we are over, we, we we're over our heel, you know, over our skis right now.

And, um, when you look at the valuations, and I think we need a correction, and I'd like to see the heat coming outta the market, and hopefully that's what's happening right now. So it's like, let's, let's get at some of the heat out and, um, maybe it won't explode if, if we do that. Absolutely.

I mean, and the other thing is, you know, hindsight is always 2020, right? com era. Hey man, I helped take a pub, a company public in that time.

We went public in early 2000, or maybe it was late. Yeah, early 2000, January, February, you know, uh, our Merrill Merrill Merrill Lynch was our lead, uh, you know, broker take banker, taking us out. We had the wonderful man, Henry Blodgett.

Henry was our lead banker. Of course, Henry got in a little trouble because he was out touting stocks that he was then selling stocks. And we were, we got wrapped up into that, right?

I'm not, we're not supposed to talk about it. I was told never to talk about it, but, um, but, um, I think the statute of limitations is run by now. It's 25 years.

com and some of these, you know, there, there was some abuses built into the system. We may find out 10 years from now, five years from now, that there are some irregularities. There are some, you know, things that we needed to correct that would prevent these kinds of things.

But I don't know if we ever prevent them. Totally. I I think it's, it's the herd.

It's the law of the herd, Right? I, I wanted to add something here because I mean, we have been talking about valuation fatigue for quite some time now. You know, that is, again, that's its real story.

And that that is happening in the market. Correction is due to kind of, uh, in due, in due course of time, we will see that. But I think there are two technical repercussions of this.

One is that, uh, the AI data center movement and the underperformance of the revenue, which is kind of coming through, and this is, uh, again, uh, showing up with, uh, what you see as a correction in the market. So there is no substantiate revenue as such, uh, to date for these AI data centers and how it'll play around. That's the first, uh, uh, fact technical factor.

The second technical factor, I would say that, uh, a multimillion dollar revenue wiped out, uh, due to the geopolitical restrictions. So what will happen as a correction is probably, and I think, uh, there is some announcement today that there will be some corrections coming from a regulatory perspective, how this chip market would be normalized or harmonized. Because if the market is reacting strongly to this, this is in making.

So these are two points I would also like to make here. I I think the geopolitical thing, Kareem, is huge. Kimberly, you had said something else out about, uh, some recent bill passed by the Senate.

Yeah, well, that's having to do with where, who's selling where and, and holding Back. Well, it's, it's geopolitical, right? They're saying, Hey, we're going to prioritize domestic customers over international customers.

You know, you see that during war time. And I read though, you said Kimberly, I'm saying, yeah, you know, during wars, we've seen that sort of strategic kind of thing. You know, in, in another time, in another place, would an Intel or an A MD or a uh, Nvidia say, Hey, you, you can't tell if they're willing to pay more money, right?

I'm not owned lock, stock and barrel by the US government here, right? And, and stuff like that. So there is, anytime you get a bubble burst and people lose money and get hurt, there's an outcry for regulation.

It's just like a knee jerk reaction, right? And, and I think we're, we're gonna see it here, assuming that this trend does, does, uh, stay the same anyway. Hey, yeah.

Substantiating this point, like 95% of the revenue for Nvidia from China is now zero. Well, you know, we, we've discussed this and we'll discuss it again, but I really gotta take us on a break here, guys, 'cause we're running late and we've got a third segment coming up. You're watching Textron Gang, Discover Textron Group, the epicenter of tech innovation.

We are your go-to for reaching IT, leaders and practitioners worldwide. Our secret impactful content that sparks awareness, engagement, and top quality leads with us. You'll access editorial websites, streaming videos, virtual events, custom content analyst research, and more.

Join our satisfied clients. Let's revolutionize your tech journey. Contact us today and tell your story to the world in the most powerful way with Textron Group.

Hey, everyone, we're back here into our C block. Thanks for staying with us. Our next, uh, segment is parsing processors.

It seems no matter what we're talking about inference and, and all of the things that could come up to displace GPUs, they are the ones present and future kings of the block of the pile. Uh, this is coming outta some futurum research. Uh, Kimberly, you want to lead us off?

Sure. Um, futurum just put out a, um, research report on the market shares of the different, um, chips or C GPUs out there, and of course, NVIDIA's at top. Um, but what is really great about the report that came out is seeing the other guys come out and, you know, why we think, see Nvidia as the big kaon here.

Um, a MD is coming up. You've got a whole bunch of other people. Um, one of the ones that I picked up on that's up there is Rock.

And I think that's because I just listened to a, you know, a discussion about Humane, who is a, a big, um, investment over there in Saudi Arabia. And their first data center is going to be based on GR and then followed by a MD. So it's kind of like they started out with the other guys as opposed to the one that we see.

But I think the other piece that you had brought up on this, um, is, are these, you know, how, how we're going at it, how this is going, you know, it's gaining ground in the data centers, um, but where is this going in terms of investment and depreciation schedules, et cetera. You know, are we doing the right things in terms of reporting, um, expenses, et cetera, that are out there, um, in terms of the, um, company. So anyway, I will turn it over to you guys to talk about this.

You know, where, and ne next week is, um, super computing. So you, you guys are gonna be right front and center where all this stuff is at. So let's talk about it.

Yeah. This, this week as, as this airs, it's just starting this week, sorry, this afternoon, this evening. Yeah, we're on Monday Through the magic of quantum time dilation here at Tech Drunk Studios.

Go, go ahead, guy. Um, so it's, it's not just competition with, uh, the kingpin of the GPU nvidia. It's competition of the, the G against the GPU generally with other processing units, so-called xus, NPUs, neuro processing units, TPUs, tensor processing units, pd, qp, like whatever it is.

Um, this has been happening for a while. I think the market is just taking time to sort of catch up to this idea. So it starts with saying, which began in graphics, which is where the name comes from.

It starts with saying that a central processing unit, a single processing unit is, may not be enough for certain special applications. Even A GPU is, is, is, uh, you could even call it, call it a generalized processing unit. It has a, a, a limited instruction set and a certain capability to it, which has been adapted very well to, uh, to, to ai.

Now, naturally, these other ones are even more specialized. What is not often talked about, um, in general, even in really technical discussions, and this is where I'm saying like the market is still catching up to this, is how many advances in chip design there have been over the last 10 years, it to be able to, it, uh, you know, more or less custom code, uh, a, a, a processor to do very specific things and then manufacture those ma manufacture them very quickly, potentially at scale inexpensively. It has to do with innovations in, in what's called packaging, meaning, you know, used to just be one chip.

Now that we have these giant chips, but we also have so-called chip lets, where you can collect a bunch of chips. It still looks like one chip. I don't wanna bore everybody with all the examples of this, but it allows for something like a software development approach to what is a physical thing that gets plugged into a board, which also has, you know, had innovations.

But the market remains highly focused. And the, the, you know, I mean, talking about technical buyers successfully being highly focused on GPUs and what the future and research report shows is not just is NVIDIA's dominance breaking, but the GPU's dominance may be starting to break as well. And, um, this is, this is really great.

That's where efficiencies come from. That's where people, of course, now the software needs to be developed in parallel to take advantage of these things. That's the next wave.

I would also like to add something here. I mean, these are a couple of things I wanted to highlight. Two things.

The first problem with the report, which, uh, Kimberly had mentioned was that the valuation of, uh, these, uh, GPU investments or chipped investments is, uh, skewed up. Because, you know, if you actually, um, have a life cycle of two to three years practically, but you are showing it in the books to five to six years, that means it's an in inflated valuation. And, uh, there, there are, uh, certain bodies like IFRS and GAP who are regulating this.

I think, uh, there is a substantial amount of speculation that how these assets would live that longer life. And of course, uh, uh, what it means to innovation, right? I mean, there is a lot of like, traction in, you know, how these chips are being developed.

And to your point, gay, that what happens is that we are seeing a surge of ps, right? And PS are like cost, uh, cost efficient and also energy efficient. So, you know, substitution towards G GPUs, but the market has speculations about it.

It's not substantiated what value it creates and all that, right? So, and how it'll be adopted. So there, there is a twofold kind of, you know, movement going on, and we need to kind of watch out how this turns around and how, how it comes across.

Because the bigger problem is that this infl inflated valuation of the assets which we have in the data center, uh, which will have repercussions that, uh, you know, these innovations, which we are talking about p and all this kind of, you know, open, open standards and, you know, software like approach to chips would be stifled. So this is, uh, the, the, the crux of, you know, how, uh, this would play around. So I, I, I have a theory.

Let me do this one, Mitch. I have a theory. This is the first generation, kind of the first wave we've seen of, of the, of the AI GPU craze, and I call it the American Cadillac era.

We, we are, we, we use GPUs for everything. Anything you, you want to do something with ai, you need A GPU, right? Because we, we have, we're overbuilding 'cause we really haven't turned the dial enough to understand what we could get away with.

And we're optimum is, we're still dialing in, is the word, right? I, I think as we go on, we're going to get better and say, Hey, instead of the GPUs, the XP, all these other things are probably more efficient give for the job that I needed to do. Not every job needs the GPU and especially as we move more from training and more to inference, which is what everyone is saying, right?

Our reliance on these Cadillacs, people are gonna say, no, I could use a Chevy for this, or a Ford or, or a Toyota or whatever you want to call it. And I, I do think that's gonna bring some rationality to the chip market. There's some good data in the report, Alan, that backs up what you're saying, um, which is the, the primary driver around GPUs is the time to train models.

Yeah. The huge investment that goes, that's what drives the biggest purchase of those. But interestingly enough, the workload is, is starting to change of what people are using GPUs and other XPS variations.

So the, the kind of farther down in the report, if kind of tuck, untuck, some of the data, uh, it talked about, uh, both training and inferencing was about 33% of the use cases followed by mostly inferencing, which is also at 33% in the same 30 range. Training dominant was really only 19. And, uh, data prep, you, the data that we use is kind of a small 10%.

But the point being is, you know, it's not just a, a linear curve or a, or a hockey sticker all built around training. Yes, that's happening. And we, they wanna drive the cost per token around down for training and make it faster, but they also need to drive the cost per token down for infantry in, in other use cases.

And that's happening. So I, it's, I think we're turning that curve that you're talking about. It's hybrid infrastructure in making, right?

I mean, if you see, uh, the XPU growth is 23%, uh, in 2026, that's the projected, uh, you know, val value of the growth. And then you see how the hybrid infrastructure would also open up new opportunities for smaller players, right? So that's another thing which we should watch out for.

If I could make, and It's worth mentioning also because, because yeah, Jack, it's just worth mentioning. 'cause Q Comms last week that DRA, which is, uh, one of these, uh, uh, Kubernetes related projects just went general availability in August. DRA, uh, dynamic resource allocation name doesn't tell you so much what it does.

It is a prime way for applications to get delivered these diverse compute resources, whether G-P-U-X-P-U or whatever. So the, the software is, is Is the platform. No, I think we, it's the next wave, but guys, we've gotta run.

Alright. You know, we try to keep these within a timeframe. Give, give Jack two seconds.

Oh, I, I, if I can just leave our audience with three thoughts, which is, I hate to be the historian here, but we've been through this before. I went through this with co-processor with, uh, you know, uh, co-processor and with, um, the, uh, risk versus SIS scores. Second thought is we do have a change in US finance laws about how you can capitalize equipment.

So we won't be seeing depreciation this year in the us which will change how all of this reports at the end of the year. And the final thought is, despite having this wonderful list of processors that are available now, GPUs, xus, CPUs, et cetera, there are two fabs in the world that make these things. There's Intel and TSMC and that's it.

And that's going to be a very big issue next year And years to come because they don't come on that quick. But speaking of quick, we've gotta quickly end the show. Mitch Guy, Kimberly Garima, Jack, thank you.

Thank you for watching. We've got text on TV following this, check it out. But this is Alan Shimel, we're out.

Hey everyone, it's Alan Shimo. Welcome to CubeCon North America 2025, or some people call it Cloud Native con, or as I called it, way too long. The line to get in here today.

They gotta do a better job. During COVID when they were checking vaccines, it wasn't this long a line. There was no reason for it.

I waited 45 minutes to get in here today. So I don't know if it's an a TL thing or an a TL cloud native thing, but we expect better anyway. Now we started early, so I had some time to kill with that.

Let me introduce you to our first guest here from the show floor at Q Con. Andy Suman. Andy, you may, if you, if you're a fan of Tech strong learning events and webinars, you might have seen Andy.

He's done more than a few with us. Um, Andy is with Fairwinds, and don't worry if you don't know Fairwinds, we're going to make sure you know him after this. Andy, welcome to our coverage.

Thanks for being our first guest. Thanks for Having me. You wait online this morning?

I did. I'm glad I made it here in time for the interview. Yeah, me too.

I was not sure I was gonna make it. It was a little crazy. Anyway, Andy, I said you're with, uh, Fairwinds, but tell us a little bit of kind of what your role at Fairwinds is and how you came to Yeah.

To that position. Yeah, Definitely. So I, I've been a long time infrastructure guy.

I've worked in infrastructures, worked on infrastructure basically since I was a kid. And, um, about nine years ago at my previous company, I got into Kubernetes, was really excited about it. It was very early days.

And then I joined, at the time, reactive ops was the name of the company. What was the name? Reactive Ops, uh, react Op after Reactive Manifesto, which many People don't know About.

All right. Uh, A little trivia right there. Yeah.

Yeah. So we, we built and maintained Kubernetes infrastructure for other companies, and I've been doing that ever since, uh, for the last seven and a half years, uh, working with Fairwinds. It's part of delivering services is an AWS partner as well.

Yes, we are, uh, an AWS advanced tier partner at the moment. Um, and we've been working closely with them for many years the entire time, but over the last couple years, we've really strengthened that partnership. The majority of our customers are on AWS uh, we are, we also operate across all three clouds, but AWS is definitely the lion's share of that and Very cool.

Yeah. So, you know, I know a little bit about Fair Ones, but let's assume the audience out here doesn't, you guys, I mean, number one, Kubernetes experts first and foremost, right? Some of the Number one thing.

Yeah. Yeah, I mean that's, I, I've listened in on a lot of the webinars. As I mentioned, the, the, the knowledge and skill gap for Kubernetes infrastructure deployments is, is second to none.

Um, but as you said, you work with all the different cloud environments. You work with a law, a big range of software, including a lot of open source tools. Yes.

Yeah. We actually have several of our own that are very popular. Goldilocks Yep.

Is something you guys have have, and it's open source, anyone. Yep. You don't have to be a, a Fairwinds customer to use Goldilocks.

And, and that's yet another interesting thing about fairwinds, right? As you guys are developing the tools that you use for your customer engagements, you actually open source them, then anyone could use them, which is pretty cool. Yeah.

Yeah. We really enjoy, you know, sharing the knowledge that we have back with the community, not just in the form of services, but also in the form of open source. Uh, I love that.

Another one of our major projects is, um, Pluto. So if anybody went, lived through the Kubernetes one 16 upgrade and all of those APIs got removed, uh, at the time it was very hard to tell if you were still using those APIs. And so we wrote Pluto to help with that.

And I think today it still helps a little bit. It's not as big of a problem as it used to be, but APIs get deprecated and removed all the time. It doesn't change.

So I, I, I wouldn't bet it against it. Not, well, I wouldn't bet against it not being a problem At some point, I'm sure In the future, again, with APIs and AI and MCO server. What about MCO servers?

What are you doing with them? MCP servers, MCP, Excuse Me. Yeah.

Um, not a ton at the moment. We do have an MCP server kind of in the works for our product that we use to do. Of course you do.

Everyone has what Cost? I mean, yeah, it's kind of table stakes at this point. Yeah.

So our, uh, software product, Fairwinds Insights, which all of our customers use to get insight about their, their, uh, clusters, and they'll be able to access that via MCP here in the, you know, within the next year or So. Of course, MCO came before MCP and I guess soon you'll have M-C-Q-R-R. But, um, Uh, Andy, for people who want to get more information on Fairwinds, where do they go?

com, uh, go to our GitHub where all of our open source lives, that's Fairwinds ops is the company on GitHub. Uh, those are the two primary ways to get ahold of us. We all have, so have an open source Slack community, uh, that you can join.

It's on any one of the read mes on our projects. Excellent. Alright.

If you don't mind, I want to kind of pivot a little bit and talk about recent, uh, development, recent announcement with you guys partnering with AWS Yeah. In the, you know, in the platform engineering space, right? Yeah.

You guys are, uh, partnering with AWS on a new IDP offering, internal development platform offering. Yep. Internal developer platform offering.

Tell us about it. Yeah, so we've always managed the base level of infrastructure, but our customers often need something above that. So it's always been, it's your job to deploy your applications and do your CICD and these days, um, I think there's a need really for, you know, platforms internally for platform teams to be able to deliver clusters and deployments and standard pa happy paths to their developers.

And so AWS has built, uh, sort of a blueprint for this with a whole bunch of different open source projects. So we've got Argo workflows, Argo cd, backstage, cross plane, wow. All put together in this really nice package.

And so what we are starting to do is we are offering this as a services offering to our customers. So we'll come in, set that up, build the, you know, kind of control plane cluster for you, give you all the patterns that you need, help you deploy your first application through the platform, and then hand that off to you. Or we can manage it long term with our standard managed services, but really it's about zero to 60 on a platform in a very short amount of time.

And so, you know, there's a few different flavors of backstage. There's open source Backstage, and then we have, you know, uh, I think Spotify still has, they have one, they productized, they, it was their, you know, they pioneered it. I'm not, I'm not banging them for it Or anything.

Oh no, absolutely. But, um, how, how does this offering stack up to some of the commercial backstage offerings? Um, I'm not entirely familiar with all of them.

You know, there are quite, there's a lot a few out there, right? There's a lot. Um, but really this is truly based on the open source.

They're also working closely with the, uh, canoe effort, uhhuh, um, to really build this fully open source. And so I think as with all open source, what you'll get is ultimate configurability. You'll be able to do anything with it you want, but it will be a little bit more complex than using an off the shelf solution.

Yeah. So it really, It's not, it's not as pretty maybe. Yeah.

Yeah. And that's where we come in. We're gonna come in and help you make it as pretty.

So I think once we are done delivering this open source to you, I think it'll stack up really well against those paid offerings. Really. Yeah.

UI and everything else. Absolutely. Yep.

Very cool. And now that's an offering you're doing with AWS? Yes.

As an AWS partner, we're working with them to deliver that. They're the ones who are they picking, like besides backstage, you mentioned that's some GI ops and some other stuff. They're the ones picking that, or does the customer get to put anything they want in there?

So they've picked kind of the core pieces of it. And then there are certain parts of it that the customer will be able to pick and choose and swap in and out, because there's an entire, um, auth solution built into it. So at the moment, the open source uses key cloak, but obviously not everybody's gonna use key cloak for SSO.

So there'll be plug and play options for SSO for your code repositories. So we'll be able to do GitLab or GitHub or Git or any of the other, you know, kind of flavors of that. Very cool.

So it'll be a little bit of both. And then because they have you doing their service, if they want to add other things that maybe even aren't in the core, right? Yeah.

We can add things on, we can help to customize it. You know, everything we do is very, uh, highly tailored to our customers. Right, right.

You know, obviously for operational reasons, we try to keep them similar. We use best practices everywhere. Right.

It's special scalability. Sure. We are a very high touch service, and we really wanna make sure our customers get the, the platform that they need for them that fits them, which is why I think a lot of off the shelf platforms don't work because they're not custom tailored, but then building your own takes so much time.

com days, you probably weren't there. I wasn't working. No, No.

I, I realized this when I was talking to someone. The other, they said, yeah, I was in high school. com days, I helped start a company that was what we call an a SP application Service provider and back.

This was before this cloud, before this hypervisor, all that stuff. Yeah. And, uh, we were offering Lotus Notes, which you probably have heard that I I've heard Of what?

Yes. Lotus Notes, PeopleSoft, Oracle, bunch of like major enterprise Onyx or CRM and stuff. Major enterprise, uh, um, applications.

Right. And the rule of thumb we learned then, it was in 80 20. No one just takes some app like that and just plugs it in and runs.

There's always about 20% that needs to be, our mics are good for that. There's always about 20 cent, 20% of customization that needs to be done. I don't think that's changed.

I don't think so either. In fact, I think the percentage might be higher when you get the open Source. Yeah.

Maybe with open source. 'cause it is, you may not have UIs and also you just have a lot more, uh, options. You know, if you want it to be green on Mondays and blue on Tuesdays with open source, you could do that.

Yep. Excellent. Yeah.

Andy, I don't know if we mentioned the website for Fairwinds. Did we? Uh, I'm not sure We did.

com. It's very simple. Very simple.

Yep. Guys, so take it from me here. Shimmy.

com. Andy, it's great to see you in person after we seeing you on those little webinar screens. Same thing.

We're gonna take a break. We're live here at Kon Cloud Native Con, we'll be back. We've got a full opening day here, so stay tuned.

Hey everyone, we are back here. This is, uh, I think gonna be our final interview for day one of Q Con and what a day it's been. You know, there's, I don't know, 10,000, 12,000 people here.

The, the expo floor is cavernous though, and cold. Very cold. I'm glad I wore a heavy sports jacket.

This poor guy's here in a short sleeved T-shirt. You're not cold. I'm moving a lot today.

So if I were sitting in a chair, you'd be cold. I'd probably be cold, but I'm moving a lot, so I'm okay. Absolutely.

But tonight is the, uh, what are they called? The cube? The cube crawl or Whatever.

Oh yeah. The coup crawl. Yeah, yeah, yeah.

The, you know, the, uh, They're gonna have drinks, food, everybody's gonna be hanging out in the Yeah. Sponsor. Yeah.

There's a word for it. Yeah. But it's cube crawl and, and it's, you know, the, the in expo hall reception.

That's right. Yeah. That's a Good word.

Yeah. Hey, if you don't know this guy sitting next to me, you probably are not a big fan of Argo or, or, uh, GI ops Octopus, Gi ops Octopus Cube. He's, he's with us almost every single CubeCon I've done.

And I probably have done, I don't know, 18 of 'em. Nine, six. Yeah.

This is my ninth year of coup con's To a year. Yeah. So you're right there with me.

I, The only one I missed was the original one in San Francisco. I missed that too. And I, ironic, strangely enough, I was like a quarter mile away from it when it was happening, doing really else.

And No, I didn't, didn't go over there. So I missed that one. But then Seattle, I think was that next one.

Yeah. And then San Diego, right? Wasn't there two Austin?

Remember where it was in Austin and it snowed? Yes, I do remember that. That's when I realized how big this was gonna be though.

Yeah. San Diego still met, well, I, I will say this. San Diego and Valencia.

Oh yeah. My two favorite ones. San Diego and Valencia were great.

Barcelona was good. Barcelona was really Good too. That might be, because Barcelona is a great, Barcelona didn't start.

It's straight. So I thought, I thought Valencia was a quieter kinder Barcelona. Ah, yeah, yeah, yeah.

It was kind of out there. It was farther away. Yeah.

Yeah. Well, there's a little bit of, like, the hype factory has kind of gone down a little bit because you have a lot of people have moved into the operations phase, right? Yes.

With this stuff. And so in the very beginning, there was such a push to like, we just gotta learn all the things and jump in and figure out what's useful and what's not. And now we're in that stage where a lot of people are like, Hey, we're operating, we're happy.

We're, we're scaling. We're, we're hitting the normal everyday challenges. I, you know, more than me on this, so I'm not gonna pretend to be an expert as you are.

But I really feel like, so first of all, at the beginning it was all developers. I felt it was very heavy developers, guys in t-shirts and backpacks. Yeah.

And Cobe was so hard still. It was really hard. Let's face it.

Yeah. It was so hard. And people were not looking for lifelines, but they just wanted to learn more, talk to people like them, do that community thing to figure this out.

I think, I'm not saying Kubernetes is easy, I'm not saying cloud native's easy today, but it's certainly not as hard as it was. Yeah. But I also think, as you said, we've moved from just pure developers to ops people mm-hmm.

To platform engineers. Yeah. To SREs, to security people.

And, and so the audience, I, I would say let's even throw in now data scientists. Mm-hmm. Right?

And, and those kind of folks. So the audience has expanded, the product mission has matured, and the products and and projects themselves have matured. Oh, Yeah.

Where I think it's, it's just a more normalized thing. I I still think the passion is there though. Oh, Yeah.

Yeah. No, and there's, there's always new stuff happening. And you know, a lot of the shift in this last couple, last two years has really been about supporting GPU workloads, AI workloads.

So there's a whole bunch of there isation happening there. But you all are doing a lot of stuff on cost savings, on upper productivity. So That's the efficiency phase, right?

Yeah. You always, first you try to just get stuff to work and then you make it more efficient. And then somewhere right after that you say, oh, we gotta worry about security too.

Um, which is a problem. Well, I, wait, we gotta stop a second. Sorry.

I didn't give you a proper Introduction. So you don't know. This is Dan Garfield.

Dan Dan has, look you pioneered Argo in a lot of ways in GI ups, right? Yeah. Um, Argo is now of course part of Octopus deploy.

Yeah. Argo as a project is maintain, Argo Is a project CNCF. Yeah.

It's maintained by Octopus deploy, uh, red Hat, Intuit, others. And so we have a partnership with a number of different companies that we maintain the project with similar to Kubernetes. Absolutely.

But I do, I do think there is a special fit because Argo the octopus, you got an orange octopus and you've got octopus deploy. We're blue octopus. So we're, we're like, I Got blue ox, Blue and orange, uh, go together, Bring me and orange.

Lemme burn you, I guess. 'cause this is the commercial entity. Yeah, that's right.

Yeah. Yeah. So I got the blue one.

But, um, and I should mention, if you don't know, and you, maybe you don't follow this closely in home, you know, um, Argo is the out of over 200 I believe projects now in CNCF Argo is number three right behind Kubernetes itself and, uh, open tell hotel. Yeah. In terms of like contributions Yes.

And activity and velocity, project velocity, yeah. Number three. And, and Oel, uh, is so generalized, right?

It, it really touches everything. So it makes sense. And, and Kubernetes of course is the foundation.

Sure. So those two make sense, but how do people deploy to Kubernetes? 60% of the time they're choosing Argo CD and number two isn't close.

It's down in the 10% range. Really? Yeah.

'cause there's a whole, there's, it's like everybody chose Argo cd and then there's kind of a sea of other things where like, they're using Jenkins and they've just always used Jenkins, so they haven't moved on from it. And they're throwing Coop detail applies using that. They may have good reasons to do so, but, but yeah.

Agreed. When we look at the marketplace, that's what we see about 60% of clusters are using Argo CD today. And, uh, and some vendors, some clouds, it's more, but yeah, we, we see pretty big adoption.

So Let me ask another question there. If you are using Argo, are you by definition doing GI ups? You are definitely facing GI Ops Now, whether or not you've actually implemented those principles, you can use Argo CD in non GI ops ways.

Uh, so for example, if you're using, if you're referencing images using floating tags, you're kind of not really full, you're not really embracing GI ops. And you can definitely, um, use Argo CD with manual syn turned on. So you don't have automated reconciliation.

So Argo CD is pushing you towards get ops. It's encouraging you to get a good do get ops, and it's the best way to do get ops in my opinion. But you can also, uh, if you wanna hold a hammer upside down to hammer in nails, you could do that.

It's just the tool doesn't want you to do that. But you could do that, You know? So I'll give you a life lesson.

As I've gotten older, Dad, The tools make the man and using the right tool for the job is all the difference in the World. That's true. Yeah, that's true.

So you, you know, you wanna keep using the hammer upside down. God bless you. But life's too short to do that.

I I agree fully. Yeah. Yeah, Yeah.

So I, I wanna get into some of the announcements. You guys have some of the news coming outta this. Yeah.

But before we do, I I, I want to kinda solidify the, so we spoke about Argo, the partnerships of running, maintaining Argo. Yeah. You are part of Octopus Deploy though, right?

Yep. That's right. And and they're though Octopus Deploy is a maintainer of Argo.

Yeah. And a large part of their business, I imagine is, you know, in, In Argo and Kubernetes. Yeah, that's right.

But Octopus Deploy in is in and of itself an entity. Yeah. Talk to us about Octopus Deploy with Argo as part of it, but nevertheless its own entity.

Yeah. So the, the history here right, is you have, the Argo project is created by Intuit. And, uh, they said, this is the way that we wanna deploy software.

There's a number of other tools within Argo that they were using as well. And they said, we really want somebody to take on the mantle of running this thing. And so they looked to us Code Fresh at the time.

Now Octopus Deploy. But, uh, they look to us to, to be that group. And so we became the first commercial vendor to come onto the project, help it get into the CNCF, help it, uh, go through the process of graduation.

And, um, to take on that stewardship as, look, we're gonna have a commercial interest in this open source project where if you're using Argo and you want to do scale deployments, ar Argo CD really has one job. Look at a source of truth and get, and get that deployed to the cluster. Everything else is kind of window dressing right?

Now, what if I want manage a change from one application to another, I wanna promote something. Well, that's where, that's where Octopus Deploy comes in. What if I need support Octopus deploy?

What if I need, um, help doing architectural planning about how I'm gonna roll this out? Octopus deploy is the answer, right? So, so as a commercial vendor, we have an interest in maintaining that stewardship of an awesome open source project.

Uh, and then providing, uh, both tools and services to help you be successful with that project. The great thing is, because we maintain the project with others, Intuit and, uh, and Red Hat Acuity, we cannot any of us take that project and go and just change the license suddenly and say, Hey, we're gonna take a bunch of features out. Uh, instead we have to focus on something that's gonna work for all of our interests.

But that's, that keeps Us on us The foundational model. It's not, yeah. You can't decide.

I've got a Secrets program, I'm changing the licensing of them. Yeah. I wanna, I wanna make money on the ui, I'm Tired 'cause I'm gonna provide a, a better version.

So I'm gonna make sure the UI sucks, you know, for the user. We, we can't, we, we can't do that stuff. We don't wanna do that stuff, obviously.

No, But, but that, again, look, I've been in the open source game a long time. That is the beauty of this foundational model where that rising tide lifts all boats. But no, no one company can, can manipulate this through their own financial gains.

It's a super important point for governance. And it also means that we have a sustainable program for development. So for example, we offer enterprise Argo support.

Yep. Now, our business octopus is really driven by selling our software. Right.

That's where we make our money. And, uh, so the way that our enterprise support program works is, hey, we're off you in surprise support, but basically that funds our open source work. Yep.

So, a as we add more, uh, enterprise support, you need, you need help with Argo cd or something goes wrong in the night with Argo rollouts, you can call us up, we'll help you fix it. Uh, if we need to provide a patch, we'll do that. But this is the way that we fund, uh, our open source contributions.

And of course, that also feeds into one of our biggest customers is the Octopus platform. Because if, if something's not working for a customer, they may not care if it's Octopus or Argo. They just want it to be fixed.

They Want one throat to Choke. Yeah. So it, so that way, um, you know, any of our customers are essentially funding these open source programs and, uh, making sure that they can be successful, which is, uh, you know, it's nice to have Yeah.

But it's, but it's really important that you have a sustainable approach to open source. Otherwise you're just borrowing somebody's time until they don't Have time decide they don't wanna do it anymore. Yeah, exactly.

Or they want to put the squeeze on to monetize. They gotta Change that license. They gotta, they gotta get, look, I saw this insecurity 20 years ago.

Yeah. You know, with, and, and look, to be fair, let me just play devil's advocate a second. Yeah.

Please. Do You get people who put their heart and soul into these open source projects? Absolutely.

You know, single company maintainer, they maintain a community where 96 or 97% or more of the people don't pay 'em a dime. Yeah. Won't even beta test or report bugs.

And it's kind of thankless. Right. And then you've got, and what really kind of, in my experience with kind of tips it is then you get other commercial entities who come by, use this open source tool that you've put your blood, sweat, and tears into.

Yeah. And they're monetizing the heck outta me. Yeah.

Taking money away from you in essence. Right. And I think for a lot of, uh, for a lot of folks, that's where that's the, the straw.

That's the bridge too far. Yeah. Right.

And so they, they do do things like changing the licensing or, you know, making it harder for other people who they consider almost like parasites. Yeah. If you, you know, I don't blame, uh, companies that need to change their license.

They're the, they're the only ones maintaining a project and they're like, look, we can't pay to fund this development. We have to, we have to find a way to make it sustainable. Totally get that.

Um, I'm really proud of the fact that we've been able to build this ecosystem with Argo, where we have multiple maintainers. Yes. Where we all have good interests.

That, Well, again, that's the foundational model. It's a cooperation model. And it's not just vendors, it's End users, it organizations into it, Into it's example.

Right. Like, they're not selling Argo services, they want you to file your taxes and manage your money stuff. Exactly.

But for them to be successful deploying their software, they rely foundationally on Argo to do it. It's so it makes sense for them to invest in it. It's key.

Uh, and then it makes sense for us as, um, to be steward good stewards of the project because ultimately that's the ecosystem that, that, uh, is, is paying The bill. Absolutely. Alright, we're running low on time, so let's turn, okay.

What's news? There's a lot of news, uh, in the Argo project. 2 just came out this year.

We shipped Argo CD three. Okay. 2.

That means all of Argo CD versions two are now out of support. They're gonna start stacking up CVEs and bugs. They will not be fixed.

A lot of people have not yet upgraded to three. This is not Java. You do not run it for 10 years without upgrading.

You move to the latest version, Uh, at your own risk. You wanna stay on that old stuff, you know, at your own risk and it's obsolete. Yeah.

And what we're talking about with, uh, the shift from Argo two to three, we made that a, uh, an arc. So we made very deliberate changes that would be very easy to upgrade through. And almost all of the behavior that's changed in Argo CD three can be changed back to two X behavior.

We have a great upgrade guide to help you do that. So that's new. That's exciting.

We've got new maintainers. We just added two additional really maintainers from Octopus Deploy focused on Argo cd, uh, who, who just got promoted on Thursday and are now Maintainers Congrat see them. Yep.

They, they put in the effort. Wanna give shout out? Yeah.

Uh, you Eugene, he was walking around here. Just a second. You have Guinea, um, as well as, uh, as, uh, pat Close, um, who wasn't able to make it this week.

Yep. Yeah. Gh what's his last Name?

Uh, den? No, not the, we have Evgeni who comes on our show once in a while, but yeah, He's, he's fairly new. But, um, we've seen these new maintainers do a couple of really great things.

Pat close, he did a big migration in Argo CD to move GI Ops engine back into Argo cd, which is a massive project. Sure. He worked with Lee Tuit to get that done.

Did a great job. 2 because, uh, the version and get changed. And he was able to track, it took three weeks to figure out, but we got it done.

So won't be hurting users anymore. We're allowed to see that. So that's going on on the, on the community version of Argo cd.

Big stuff happening there. Uh, and of course, as I mentioned, we offer our enterprise support for Argo, our technical account management, where we do proactive stuff, but we also just shipped a lot of new features in Octopus to improve your experience with Oh, very cool. Argo.

So if you wanna stage out, you know, uh, for example, let's say I've got 10,000 storefronts. They've each got a Kubernetes cluster, they each have an Argo instance, and I want to orchestrate promoting and managing all those Argo instances. I can do that with Octopus.

If I wanna manage deploying new versions of my software to all those versions, I can do that with Octopus and That. And that's the, so historically, that's where a commercial company that's maintaining an open source project makes their bones. Yeah.

You wanna scale to that level. It's hard as hell that you probably could do it at some level using just the pure open source. Yeah.

You can write a lot of scripts, a lot of glue, but you've got a great foundation with cd. She's why you do It. Right.

You know, it was the same thing. Remember back in the CloudBees Jenkins days, right? Yeah.

CloudBees knew that when you ran, I forgot what it was, four, four instances of Jenkins, you were probably ready for Cloud B'S enterprise. Yeah. Right.

Because that's the scalability that I think are in there. Yeah. And it's similar.

I mean, you could, of course you can manage. I, I know people that have 50,000 Jenkins instances and they're all 10 versions behind. Well, there Is that because they, they didn't take a proactive approach to managing it.

And that's, that's difficult place to be. Uh, but yeah, there's the not only scalability, but usability and, um, user experience, things that we've been able to add. So those are new features that we've brought in to Octopus.

Now for those that remember Codefresh, of course Codefresh is still running ci and we've got these other, uh, components, GI ops things that we're doing. But, um, bringing in a lot of the learnings from that platform into Octopus, uh, as a unified platform is something that we're doing right now. And we just launched that, uh, very cool for OpCon and, and people, uh, have been using it and building on it and growing with it.

And it's based on a foundation that's, you know, a decade old at this point, because it has the ability to do all of your traceability and your, uh, governance, tracking, you know, compliance stuff, as well as all of these scalability features. So it is really nice. It's a kind of a peanut butter jelly.

Best of both worlds coming together. I love it. Situation.

Dan, we're almost outta time. Two things I need you to tell 'em. Number one, people who wanna follow with Octopus Deploy and what's going on there?

What's the website? com Number two, Argo people who are into the, the project, they could go to GitHub, they could go to the CNCF. Yeah.

Or they could go to Argo unpacked. This is our new Argo podcast. We've done, I think, really episodes now.

We have 1500 subscribers. We just lost it. Oh, it's a big on YouTube.

We have 1500 subscribers. That's beautiful. Uh, which is awesome.

And of course you can subscribe to that on your podcast app. Favorite podcast, Argo unpacked. We talked through technical issues.

We talked through strategic issues, cultural, philosophical as it relates to delivering software using Argo CD using Argo rollouts. And we love it. Uh, yeah, we're loving that new show.

Alan, appreciate you Mrs. Garfield. He did a hell of a job Here today.

I'm awake this time. You have a wake this time. We're Proud of you, Dan.

It's good to have you. Thanks, Al. Good luck.

Continued success and keep doing what you're doing, man. Thank you. Appreciate it.

Dan Garfield here on Text Drunk tv. We're gonna take, well actually that's gonna wrap up day one. We'll be back tomorrow with more.

I've got some parties to go to. This is Alan Shimmel, we're out. Hi, uh, my name is Toma.

I work for the, as a program manager, uh, of the development tools at the Lys Foundation. Uh, it covers a few different topics. Uh, first the Eclipse id, uh, the one most of the people think, uh, of when they hear about the Eclipse FSE Foundation, most of the time, uh, it's over 20 years old now, uh, but it's still widely used.

Not so much as an id, but much more as a platform. And that's why we, we still, uh, take care of it, uh, through for a working foundation. Uh, then I am also in charge of the OpenX Working Group, which is basically, uh, the, the only open vulnerable marketplace of VS code extension.

Uh, this is a good alternative to, to, to the M Microsoft vs code marketplace. And finally, I was also in charge of the cloud dev tool working group, uh, which is the place for, for the new generation of open source tooling project. Uh, it's more than 15 projects.

Uh, most of them are using, uh, modern technology web technologies. And I will focus today on two of them, uh, eccl and AI and, and both are really fast growing and very promising projects. So, so, so let's start, uh, you know, in my role as a cloud foundation, I, I speak with many developers, of course, all the time, and, and they're working for project in the industry and also from DevOps teams.

And when I ask, uh, the question, uh, do you use AI assistant for coding? Uh, the answer is yes, in something like 50 to 80% of the case. But of often it, it comes with, uh, little hesitation and, and or smile or something that says, okay, but in my company, it's, it hasn't really approved yet, or we're still tr struggling with, with that.

And, and that's exactly, uh, the starting point of this talk. Uh, AI assistance are already part of the everyday, uh, reality for developers. Uh, even if some organization are still catching up.

And of course, it, it, it come because it come with new waste. And we with new limitation. Uh, one year ago, uh, we, we, we had this study from, from McKinsey, uh, based around AI and coding assistance.

And the prediction at that time was that by 2 28, uh, 75% of enterprise software, uh, engineer will rely on AI tos, uh, assistance for coding. Uh, honestly, my guess is that, uh, probably if we run the same, uh, study today, those numbers will be like, will likely be higher. Uh, but the real question is not adoption.

Uh, it, it's, uh, the integration and how we bring these tools into teams and pipelines in that with the right gateway and things like security and data boundaries and Providence audits, it's, it's where we are going here. Uh, no, let, let, let, uh, make you guess if you, if you are already using, uh, AI coding assistance, uh, it's probably one of the tool on the screen. Maybe I, I forgot some of them, but just sharing most of the popular ones on gig co-pilot, so and so on code, also known as Windsurf and so on, they are really powerful tools.

No discussion about that. Uh, but here is a challenge. Uh, most of this tool, if you look lists are ever closed, proprietary, or if even if they are open source, they are, they are tied to a single vendor.

And, and that's the case as a vendor. Uh, in that case, the vendor decide how the AI work, where your code is sent, and how much control you have over your, your own data. And it brings, uh, us two, two very important, uh, questions.

And, and this is a question that people sometimes prefer to, to, to avoid, uh, uh, even if you are coding for, as a hobby or, or working in a highly regulated and sensitive industry. And here, there are, uh, two simple question actually. Uh, first, where those my cut goals?

Uh, e even if you're on, yeah, as I said, even if you're not working on top secret government software are very sensitive, uh, you might be building the core algorithm of, of your next startup maybe. Or, um, maybe if it's the business critical, you, you probably don't want that code handing up in, in, in and feeding the AI model and handing on someone else screen. And the third question is what I am putting in without knowing exactly, uh, because when you generate code with AI assistance, uh, it could be called linked to a patent, or it could be, uh, something under a receipted restrictive license like, like JPL or, or any other.

And you may discover, um, probably too late that you, you, you are, uh, when you are ready to ship your product that you, you, you cannot actually use it. And these are not h cases, uh, there are real risk, and probably it'll only grow as AI becomes a normal part of your coding workflow or DevOps workflow. And as code moves faster and faster, um, this is something we, we, we are really taking this.

Yeah, this is really to, to something to look at probably. Yeah, I will be very quick on this one. I think people are probably are really convinced, convinced, uh, about why, uh, and what are the drawbacks of, um, of proprietary on signal vendor?

Uh, tools. First, the security and compliance and many, uh, of these tools require, require something large part, uh, of your code, sometimes you entire code base, uh, to, to, to a model, to something to, to somewhere else. And in some industries, of course, that's an immediate compliance broker.

Uh, for, for DevOps teams, it could, can also raise questions about how code flows through pipelines environment. The second agreement is, uh, uh, I would say custom visibility. And developers are used to, to tune their tools for their workflow, their language, their specific domain needs.

And with CLO cross, uh, tools, customization is most of the time very limited and, and sometimes simply not possible. Um, third, vendor locking, uh, a failure. This one is very, very now well known.

Uh, the way I see, it's that when you outsource a big part of your, of your daily work, uh, to one provider, you also give you, you also give control over your, your productivity. Uh, a change in pricing, a change in licensing, or even an outage can leave you completely stuck. And fourth, uh, innovation.

And, you know, maybe the most important one here, uh, you know, for four decades, uh, innovation especially, uh, in the development tools, um, has come from open source and property tools take, basically take that away. And, and with it, uh, the ability for communities to, to, to build and to innovate, uh, and to, to make something bigger. So this is exactly, uh, where, uh, uh, the problems that, uh, tier and the project, uh, would like to, to to present you today is, is designed to avoid.

So what is thi uh, basically I at the bottom, uh, this is, this is a platform, this is a framework. Uh, tier is a platform. I, I, as you can see on the screen, it's a very, uh, it's very successful, as I said, very promis, fast growing project.

Uh, more than 20,000 stars on GitHub and more than 2005 hundreds. This is an active and large community, uh, and, and by the, and that's why if you have ever, uh, wanted to join something, uh, meaningful in open source, uh, this is really, this is really a great opportunity. And, and fear ai, uh, brings AI capabilities directly into this platform.

On top of this, you have, uh, as I said, thet platform allows you to create its, uh, browser base and desktop base tooling. And not only Id actually, but the community, community, uh, decided to, to develop, uh, an ID called tier id. And it's a, it's a complete ready to use.

And if you are familiar with, uh, with the corp, for instance, you, it'll, you will find, uh, really a sta state of the art. Uh, id, uh, people most of the time will install a tier for the first time. They even say, oh, it, it looks like it's VS code, but actually it's not.

It's using the same ter, for instance, and or even sometimes have the feeling that it the same tool, but it's, no, it's not a VS code, it's not even a fork vs code, meaning that we don't have the constraints of its folks. By the way, it's a really open, free to use, uh, open source and community, uh, effort. And so, let give you a bit more detail about tier and how tier ai, uh, ID, uh, uh, breaks, uh, to, to, to, to a developer.

Uh, and what is really important here that everything is, uh, open and transparent. And, and this is the most important, uh, thing, what is the most important differentiator of, of this project. But if you look at this, I, I I'm sharing on the right, you can use any LLM, you know, the LLM, it's basically, it's your, the brain of your AI system.

You can decide the model you use and how you use it. You can use it locally, remotely, cannot hybrid. So if you want to completely disconnect your system from the internet, you can, it's very, it is a very important, uh, uh, set for, for people working in some, uh, confidential or sensitive industries.

Um, you, you can, you can go hybrid of course as well. If you, you, you need to mix your use case beyond that, uh, there is a point editor, which is based on open source library. You can see, edit and share your prompt.

This is a community effort, you know, uh, in if you, if you have ever switched between LLM, if you are family, you are with prompting and you, you know, that you often need to fine tune your request depending on the LM you use. And if you rely on all the vendor on proprietary tools that I, uh, before I mentioned before, uh, you, you, you cannot change anything here. You rely on the vendor to do this.

And here, this is done by you, or you can tune it and you can rely on the committee to, to, to adapt to your context and to your, you are using, of course, it comes with very, very, uh, state-of-the-art features, like having a agent. Uh, and you can create your own agent and you can also connect with external system. And I will give you a bit more detail about that.

Now, getting agent first, uh, yeah, you, you, you have, uh, uh, you have a set of read use, uh, agents, uh, like coding agent and testing agents or an orchestrator to manage a direction between agents. And you can also develop your custom agents. Uh, it's good to know that, uh, you, you can add a new agent, you mission based AI assistant, basically, uh, ever by coding, but can, and you can also add them, uh, using a new ai.

So if you are not, if you don't want to, to spend too much time, uh, uh, working complex tasks to, to complex tasks, to, to develop your own agile, you, you also have, uh, ui, uh, uh, UI features to, to do this. Uh, and of course, uh, when you develop new agents, what is very important to be able to interact, to integrate into your, uh, workflow and your pipelines. And, and this is where, uh, the model context protocols, uh, comes in.

Uh, it gives agents, uh, more context and direct access to po powerful external, uh, capabilities. Uh, so yeah, I think this is very, very, uh, uh, common, uh, feature of AI system today. You have thousands of, uh, NCP server everywhere.

Uh, you can connect to, uh, gate GI chat systems like Drive Google Drive or any, you have many corporate and, and external system and services, uh, now available. And, uh, where, where you can integrate. And then you basically, when you are using a TR, uh, id, or even if you are using the T framework to build your own tool, you, you'll be able to configure, uh, really easily, uh, the connection to any n CT server.

And it'll, uh, basically tell, uh, let the LLM know that you have, uh, that this new capability, this new external system exists. To give you an example, you connect it to a GitHub, uh, service, and then you can create an agent that would automatically create, uh, GitHub issues when, uh, the testing agent, for instance, uh, detect feedback. And you can imagine whatever use case you want with that.

Uh, so with that, the, you are using as now the capability to directly, uh, get the result and, or act and take action, uh, through the, so with, it's very important to, to mention that feature because this is very, the way you integrate your, uh, your features on your, your basically, you know, when you are developing DevOps feature, you, you, you have a pipeline and you need to have this interaction, this integration and tuning capabilities. And now of course, it's a very complex slide, and I won't give you, I want this one Don worry. I won't describe all the boxes here.

Um, what, what I have shown, so used so far is the id, uh, here, this is an architecture drag the description of the I platform, the tier framework. Uh, as you can see, TII framework comes with many other features build, and there are building blocks. And of course, everything here is, uh, again, it's very important and everything is fully open with a license, which explicit, which is explicitly designed to allow, uh, commercial use that that's intentional, right?

Uh, we want, we want people, uh, in this open project to build on top of it and to bring their own tools into this ecosystem. Uh, and basically that, that brings, uh, me to, to, to, to lastly stage of the day. Um, I, I already, yeah.

You know, if you look at the, at the market and the tools that we have today, I already think we, we are, uh, at a crossroads. Uh, AI in deliverable tools is exploding, but really here we have a, we have a choice to make proprietary urban lock or, or versus open or community driven. And if you look, uh, the past two 20 or 30 years, uh, the, the tools, uh, uh, that shaped our industry, uh, succeeded, uh, because they were open.

Uh, uh, and so this is exactly, uh, why I really, uh, believe that's, it's ai, it's quite project. And rather, I'm making this presentation today, uh, coming with freedom, transparency, innovation. Uh, it's, it's, it's, I would say it's time to act now.

It's, it's now because to you, you have so many tools and, and so, so many innovation, so much tools coming, right now it's time to, to go for open, uh, endpoint source ecosystem. So here is what I would like you to, to, to keep as takeaways, uh, issue. First, as a developer, a really easy, you can try the ai, uh, fee.

ID it's, as I say, again, it's free, it's open source, uh, and give feedback. You can adapt to your needs. And if you want, you can please feel free to contribute, uh, several as a company.

Uh, you can evaluate, uh, fee ID as an open solution, uh, where, where you can really have a full control. As I said, I, I just mentioned a few use case like having local LN and so on. But it's, uh, most of the time very, uh, critical need, uh, and keeping control of your tools, your models, your data is, yeah, I think something, I know some, something which is not optional.

And third, uh, as a tool builder, uh, it's not only as an an id, as I said, uh, using Eclipse tier and the ai, uh, uh, is, uh, is a way to create your own, your own tool, uh, your specific tool for your domains, for your user or for your specific workflows. So basically, whatever your role is, um, yeah, there is a way to to, to be part of, to be part of it. And remember, uh, this is, uh, the time is now.

So that was my last slide. I'm ready to hear your question. Um, you have the links, you have a way to connect with me, so I'm listening to you.

Hey guys, thanks for the tour. We're here with Yasmin Rabi, who is COO of Cloud Vault. And we're having a little chat about, well, whether or not Kubernetes is the defacto default platform for AI or not.

Yasmin, welcome to the show. Thanks for having me. We have seen at least early on that the hyperscalers pretty much use Kubernetes extensively, but in the enterprise, things may be a little bit different.

Some folks are using Kubernetes, some folks are using other platforms. What are you guys seeing? Um, I mean, for us, obviously we're biased 'cause we get involved if an organization is, uh, using Kubernetes.

And I think what I see the most is that across every enterprise, there's at least a little bit of Kubernetes. It depends on whether there's a lot or, um, just, you know, in pockets. Uh, where we see the most Kubernetes is like fintechs.

Uh, large banks are mostly deployed on Kubernetes, and then a lot of hosting platforms as well. When you have multiple copies of the same app that needs to scale differently for each customer, uh, we'll see a lot of those types of organizations on Kubernetes as well. Yeah.

In theory, will AI drive more adoption of Kubernetes? Because from what I can see, there's a lot of data processing and it needs to scale up and scale back down. And it seems like Kubernetes is the natural orchestration engine for that.

But are there other ways of skinning that cat? Um, there's definitely other ways of skinning the cat, but what I have seen a lot is, uh, for example, spark jobs on Kubernetes has been exploding. Uh, 'cause people have a lot of the data jobs they run for like, from anywhere from five seconds to five minutes.

Um, and whether or not they should be on Kubernetes, we're seeing a lot of that on Kubernetes. Uh, and I do think it's driving the footprint in the organizations as well. Mm-hmm.

Is there gonna wind up being a Kubernetes skill shortage as a result? Most likely. I mean, we kind of are at one, uh, even now, uh, a lot of the organizations I talked to, they'll have like one or two experts, and then everyone else is still in that early learning journey.

Um, even most recently, the CNCF came out with a, like a Kubernetes AI conformance document, um, where they're starting to lay out what, what are the best practices for being, uh, AI conformant on Kubernetes. Um, and even just to get there, like you need to be Kubernetes conformant, there's some standards and things you need to hit, but, uh, what they launched was essentially a way for people to do their own kind of self-assessment of do I have the must have practices in place? Do I have the should haves?

Well, obviously, like we'll see that evolve, uh, over time in the next year. But, uh, it's giving people a place to start because I think a lot of folks were like, I don't know what to do. Like, the question we get a ton is, okay, I'm starting to do AI on Kubernetes, but like, how do I optimize this?

Where do I start? And, and most folks don't know where to start. I think part of the issue with Kubernetes has always been that there's too many knobs to turn.

So can we just manage this at a higher level of abstraction where mere mortals can do this and not everybody has to be a certified Kubernetes engineer. That's a lot of what we try and focus on is what are the pieces, what are the knobs that maybe you don't need to tune and you can let a machine tune. Uh, it's funny 'cause we're, we're an ML company and often that gets grouped into like AI ml and it's like, oh, do you use AI to then optimize ai?

Uh, what I like to tell people is like, what we're doing is just advanced math. And if you can use math to solve problems that humans don't need to, for example, configuring Kubernetes. So how do you set your requests?

How do you set your limits? How do you do that for jobs that only stick around for a short amount of time, but when you do them at scale, that starts to take up resources, that type of abstraction for humans, kind of take some of that toil out of it. And then the, the thing that has, uh, helped us through for decades now is just automation.

So rather than having a human go manually configure things, deploy things more with a push button style is just have the automation run it, um, and, uh, run at the intervals that make sense to the organization, but use automation to solve the problem. Mm-hmm. We've been having this conversation about stateful versus stateless on Kubernetes for as long as I can remember.

Does AI kind of result in a lot more stateful applications running on Kubernetes itself, or is the data being access externally? Um, I think I, I'll say it's, I've seen a mix of both. Uh, we still are seeing a lot of stateful apps, uh, on Kubernetes.

And, uh, whether or not people are addressing the best practices, I think comes to the maturity of the organization. And to your previous point, the skill sets they have in Kubernetes. Um, I think no matter what the technology is, there's always the right way to do it, and not everybody follows the right way to do it.

So you kind of have to build the tooling around anyway that a, that an organization might choose to do it. Um, as you kind of think all this through for it, where are you seeing Kubernetes? Is it all up in the cloud or are people deploying it on premise?

Because a lot of these AI apps are accessing data that already exist locally and maybe, I don't know, just data gravity just driving this whole decision? That's a great question. Honestly, across our, uh, customers, it's probably a split.

Maybe it's like 60 40 public cloud to on-prem. Um, but we, uh, a lot of folks are doing Kubernetes on-prem, and they'll either do vanilla Kubernetes or OpenShift is, uh, I'd say in the last six months I've heard way more OpenShift than I have probably in the last two years. I, I think the, uh, migration off of VMware is, OpenShift is a nice, uh, first step to, for folks that are moving to Kubernetes before maybe they go, uh, either vanilla Kubernetes or, uh, go into the public cloud.

So, um, we are seeing a lot of on-prem Kubernetes as well. Yeah. What do you see people doing that you just shake your head a little bit and say, folks, maybe we should be a little bit smarter than that.

I mean, I'm, I'm definitely biased here, but like, people setting requests, like manually setting requests. So our tagline at CubeCon, it's always on our booth, and sometimes we're like, should we change this? But it still gets people, it still gets the conversation going.

So it literally says stop setting requests. And obviously, I don't mean like, don't set your requests, you should set your request, but stop sending it manually. Um, and that is something I still, because most deployments are in the, like we have customers that have millions of containers that are spinning up and down every day.

There's no way a human can do that manually. So, um, and if you do, you're probably gonna then vanilla spread it across everything and set an average maybe a P 95. Um, so that is the the biggest thing I'll say of like, people should stop setting these things manually.

You talked about predictive AI and using that to kind of manage the cluster, but will people also start using more gen AI tools themselves to manage the Kubernetes cluster and hopefully maybe something like an AI agent might explain how something actually works? Yeah. Uh, it's a great question and something where actively working on and, uh, I'll say arguing about as well.

'cause um, we, a lot of our customers have been like, yeah, I would love to use a chat bot, for example, to interact with the software and not, you know, I don't wanna teach my users how to go in and make changes or use annotations or anything like that. Um, but we're, um, some folks are drawing the line is if it's a read action, so like, tell me about my infrastructure. Tell me the last time a recommendation was deployed, how much have I saved?

That sort of thing. Cool. But if it's a right action, they're like, I don't want the chat bot anywhere near that.

I don't want gen AI to then go make like real changes in my infrastructure. So it'll be interesting to see how it plays out for folks from us as a design principle, we've always said like, minimal amount of permissions, and then you can always add on if you choose as the user to opt in. So we'll still give people access to do things, but we won't have that as a default.

So then the, the folks, the, uh, IT guys that are a little bit more conservative can kind of have control over how, uh, either agents or chatbots then interact with the software, Right? Because I don't think we fully solved this hallucination issue. So even if it's right, 99 times out of a hundred that one time is wrong will be the most mission critical application in the stack.

Right? Exactly. And the most important thing in, in everything we do is trust.

Once you lose trust, whether it's ai, ml, it's just basic automation. Once you lose that trust, it's really hard to gain back. And especially for the folks that are deploying software in their environment, they don't want their own internal users, the developers to lose trust in them.

So no one wants to get paged in the middle of the night. Um, and we don't wanna cause that to happen. Do you think at some point we might have AI agents that are monitoring the activity of other AI agents and that's how we're gonna have layers of things to kinda, um, minimize any disruptions as much as possible?

I mean, sadly, yes. I do think we're headed towards a kind of crazy landscape. Um, we'll, we'll see where we actually end up, but I like, I think sadly that is a possibility.

Mm-hmm. Um, as we kinda look at this whole thing, is there an opportunity maybe now as Kubernetes goes mainstream to just say, um, can we simplify this in some ways at the, and maybe that's a job for the technical oversight committee, or is there another way to think about this, where we just wrap Kubernetes in a bunch of things and there's a bunch of different projects because the TOC community doesn't seem to wanna get involved in that, the management of Kubernetes, if that makes sense? Yeah, It, it does make sense.

I, I think we'll see a mix of vendors that try and solve the problem and the CNCF, uh, at the end of the day, like their resources are limited, so they're trying to do what they can with what they have. And I do think the, uh, AI conformist document is a great first step, um, in this space. And we'll see in 2026 how much of that goes from, uh, like a self-assessment that is optional to more required things that should be in place.

Um, but it's also opening ground for a vendor that has maybe, uh, some, you know, investment, uh, behind it to come in and provide that abstraction layer on top of four challenges that come up. Mm-hmm. It also seems like maybe we're starting to see some separations of concerns here, because early on there was a data Science Tiger team, and if they were lucky, they had somebody who knew something about infrastructure, and more often than not, they were unlucky.

And the data science people ran into the same issues with Kubernetes that developers did. They said, this stuff's hard. I don't know how it works.

And it's very complicated. Is the inference engines though, or now being separated from the training exercise? And so, um, is that inference engine gonna be something managed more by a traditional IT DevOps team and that's gonna be part of the Kubernetes stack and that's how we'll go forward?

Yeah, for sure. I mean, that's how we do it here. Uh, I'll say formally Storm Forge, um, of course the, the PhD machine learning guys were the ones that came up with the algorithm, um, and kind of set up both the training and inference engines.

But now that we're kind of in a run IT mode, and especially for all, all of the data we look at, uh, each workload gets its own unique model. Um, and so we need various sorts of ML ops, uh, for our inference engine, and it's all run by the, uh, SREs. Uh, sure the machine learning guys get involved sometimes, um, but a lot of that is actually managed, so all managed on Kubernetes and it is managed by the infrastructure team internally.

And we've seen just, uh, for Storm Forge, we've been around doing like Kubernetes with machine learning for seven, eight years now. Um, and that skillset, uh, is, there's not a lot of, uh, ML guys that understand Kubernetes and vice versa, but we found it to be critical for being able to do what we need to do, and I think we'll see more of that. Mm-hmm.

So you've been at that for a little while now. What do you know now, uh, after having gone through all that process that you kind of wish you knew when you first started? Yeah, so, uh, I will, you know, obviously credit for the team, they've been doing this for much longer than I have.

Uh, I think I've been doing it for three, four years now here at Storm Forge. Um, but one thing that we all talk about of we wish we knew in the beginning is you can have the most advanced technology, the coolest machine learning algorithms. None of it matters if the software isn't easy to use.

Um, because at the end of the day, the people who are hands on keyboard have a lot of things they need to do in a day. Um, and learning how to administrate and manage your software is not top of their list. Um, so I think the biggest learning for us, uh, luckily we, we learned that a couple years ago and, and got to, you know, put that in, uh, to the product.

But it's something that is a design principle for us is that it doesn't matter how good the technology is, if it's not simple to use, it doesn't drop in within the ecosystem. Technology won't be used. Hmm.

At the end of the day. You guys are really at the forefront of automation in general and have been doing this for a while. What's next when it comes to IT?

Automation? What can we expect? Yeah, uh, so what's really cool, uh, now, you know, being a part of Cloud Bolt is that the portfolio has expanded.

So, uh, cloud bolt's, cloud management platform has been around for like 15 years now, doing every type of infrastructure. So being able to take our, uh, machine learning and infrastructure automation kind of technology and, and address that across everything is super cool. Um, and then vice versa, right?

Like, uh, the CMP platform today does everything from zero to one. So provisioning, patch management, um, ongoing day two operations, and taking Kubernetes and putting that part of it in all the, the pieces that you need to do, like maybe you have to set up a, an OpenShift cluster, for example. Um, now we can kind of take a step back and look at the entire life cycle, which is really cool at spart of Cloud Vault.

Um, and then, uh, we have visibility into all the billing data. 'cause we have a finops product as well. So, one thing that we will be sharing at, um, at CubeCon, and for anyone interested who wants to see it, they, they should totally come by the booth.

But, uh, we are collecting the billing data for all the Kubernetes costs and then mapping that into your usage. So you can see how much does this black box actually cost me? How do I split that by labels, by different teams?

How do I wanna share my, uh, all the shared costs, whether it's like idle costs, it's cube system, the things that people struggle with, um, being able to break down that visibility and then, um, go across your Kubernetes estate, uh, is kind of where we're headed and, and what we're working on right now. Uh, um, if you were appointed the, um, Monarch of Kubernetes, puts that one thing that you would decide that we all need to do to change the platform or fix tomorrow, that would make everybody's life easier. You have total control.

What are you gonna do? Well, control, I would, uh, get in place pod resizing to ga. It's the, the thing that everybody asks for.

Um, not all applications, even if they should be able to take restarts, can take restarts. And, um, it, it's currently in beta, which is a little bit of a tease because it works in some scenarios. Doesn't work at all.

And, um, I think that'll be a big game changer for people. All right, folks sharing it here. AI workloads are coming to Kubernetes, there's no doubt.

The only question is as well, just how challenging is that gonna be to manage? And hopefully it's gonna get simpler. Yasmin, thanks for being on the chef.

Thank you for having me. All right. And back to you guys in Steve.

Hey guys, thanks. Fed Throw. We're here with Emilio Salvador, who's vice president of strategy and developer relations for GitLab.

And we're having a little chat about how AI agents are gonna evolve in the software engineering workflows that we're about to build and deploy. Emilio, welcome to show. Mike, thanks very much for having me here.

It's a true pleasure, Right? I think everybody's kind of excited about the idea of AI agents, but we're a little bit perplexed about how this all might play out. And I put this to you this way.

So let's say every developer has one, two, maybe 10 of their own AI agents, and then, um, they're gonna have a bunch that they're gonna have to interact with other developers who have AI agents. And then there's also gonna be AI agents that will be assigned specific tasks on behalf of the team, and they will be monitoring those tasks, and then somebody has to manage all of this. So how do you kinda see this all playing out?

Uh, it's, it's a great question. And, um, I think that we are all learning along the way. Uh, at Glab, we believe that the future of software development is going to be in a space where both agents and humans, um, incorporated with each other.

And what we, we believe that the role of developers, uh, is going to change dramatically. I think that in the past, what we used to see is that there used to be a couple of tools, a couple of frameworks, but now with the rise of ai, I mean that that entire environment is exploding. Every day there's a new tool, there's a new agent, there are new capabilities, and in many cases is that you are even concerned about making a decision because you're just waiting for the next thing that is about to happen.

Every, every week there's, there's new release. So the role of developers is changing in the sense that the, as you explained before, it's not going to be a, a solo job anymore. Developers are gonna become more like architects, like more like team, um, team leaders, where you are living a team of agents that will do work on your behalf and are gonna allow you to focus on the things that matter the most, which is about solving problems.

I feel that we all know, uh, that developer or coding is just, you know, a third of the job trying to solve that problem and trying to identify how you use coding solve that problem is a big chunk of that, of that work. Now, developers are gonna focus more time on finding ways to solve that problem and using agents to delegate those tasks. The other thing that everyone who is using AI today realizes is that, is all those agents are not perfect.

I mean, that's like any, any other team member is that you have amazing team members. They are, you have team members that are really good at something, but are not really good at, um, many other things. And you need also to understand those technologies, how they work, how they play a role into your software development life cycle, and find ways to use them effectively and find ways to change them.

So the role, uh, will also change in the sense that we see more people acting as the way we call them, you know, the, the guardians of, of your code, the AI guardians, those who, those people who will guarantee that the word that is being done for those agents meet your expectations or your company's expectations from a security, from a quality, from a compliance standpoint. Mm-hmm. And as we see those agents evolving, um, uh, what we believe is, uh, there will be a rise of what we call this idea of meta agents, which is, um, think of it, an agent of agents that will play a more proactive role across your entire software development lifecycle.

It is no longer about you asking an agent to do things for you. It's an agent acting like a true team member. That agent will have an email account, a slack account, probably a phone number, an email address, and you'll end up interacting with those agents the same way that you interact with humans today.

So how will these agents kinda negotiate with each other? And I asked the question because application development, whether it's done by humans or agents, is gonna be a team sport, and different agents will be doing different things. And how will they, um, communicate with each other and kind of determine who's gonna do what, are they just gonna call us to figure it out?

I think that, um, that, that is a great question, and that's why I believe I'm super excited about the role that we can play in, in that, in that new world. Because at the very end of the day, what you are gonna need is that a platform that will enable that collaboration between either human agents or agents with agents, as you said. I mean, development is a team sport.

And now the difference is that in the team, we're gonna have different team players. And a platform like Glab will help. I mean, those interactions to be coordinated in, in a way that deliver business results.

And at the same time, I also think that we might be looking at a world where there are agents that are being deployed to verify and validate the work of other agents. And we're gonna have this kind of layered hierarchy of agents that are kind of working together, but we can't bet everything on one agent or for that matter, one LLM. Um, not really.

I think that as we see the rapid change in this industry, it's not that we are gonna see, I mean, different agents. It's that we're gonna see like different models specialized in different things. And we'll see agents basically overseeing the work of all the agents.

But I don't see a future where humans will not be in the loop. Eventually, there will be a mer request, there will be code that will need to be validated with the help of AI by, by a human. I don't see, uh, enterprise software being deployed with no human interaction whatsoever.

I can't imagine a bank in the US or in Europe deploying a new version of the software without being validated by the existing software development team. They will be able to innovate much faster thanks to platforms like GitLab. But the role, uh, the role of the mechanics of how software gets developed, validated, tested, uh, undeployed will not change.

Yeah, I mean, to your point, I don't think we can hold AI agents legally responsible for anything. And we can't say that the AI agent ate my homework, right? That's correct.

Is that, can you imagine that there was a, a, a, a big problem in the bank and someone says, well, it was not fault, it was, I mean, agent XI don't see that happening. Mm-hmm. What's your best advice for folks about how to get down this path?

'cause sometimes, you know, folks will say, uh, luck is the residue of good design. So what should we be doing today to kind of enable this future tomorrow? Um, well, I think my, my advice number one is that embrace the change.

I mean, um, this is, uh, I feel that, uh, an unstoppable embrace the change and find the tools that you fill are the right ones for the job and get familiar with them. If you look at the latest, um, surveys and studies that have been published lately, what we are seeing in, in development is unlike a year, a year and a half ago, we are now seeing productivity gains, significant productivity gains. And yeah, it's true that the technology has evolved, but the reality is that the main change has been that developers are now used to, uh, basically using those tools.

They know how to use them. We are moving away from the hype where AI was able to solve everything for everyone. And now we are coming to realize that AI is an amazing tool, but it's a tool unlike any other tool.

It needs to be properly used. If you don't use it properly, it may cause, you know, uh, some damage. Mm-hmm.

So, embrace change, try new tools, and, uh, of course unbiased. I, I, I believe that platforms like glab will help you navigate that change. Keeping two ideas in mind, extensibility and interoperability, which are the fundamentals of the new duo agent platform.

Mm-hmm. What would be the impact on DevSecOps? And I'm asking this question too, 'cause it, it seems we went from, uh, being afraid of AI tools to now everybody's using those AI tools, but now we have a shadow AI tool problem, and everybody's not quite clear what they're using those tools for.

And the output can vary greatly. Yeah. And I'm, I'm, I'm glad that you're asking that question because precisely everything that we have been working on is based on that, on that, on that question.

The idea of as, as these new technologies, new agents, new models, we need to find a way to bring all those together in a way that don't compromise the way you, I mean, develop software. And that's what we're bringing with GitLab and the dual agent platform. So you can use the tools that you want, you can use the models that you need in order to, uh, achieve, uh, the work that you need to deliver on, but keep that in a control environment.

You need a platform that will bring you full visibility on what's happening across your entire software development, life, life cycle, and will maximize your opportunity to deliver innovation faster. Mm-hmm. Do you think early on we may be obsessed with just using the AI tools to write code, when it seems to me most of the bottlenecks that exist in our DevOps workflows have very little to do with the writing of the code.

And there's all these tasks that we need to do today, and there's a lot of manual effort. It takes the joy outta software development. And frankly, for a lot of folks, the fun part is actually writing the code Could not agree more with you, but not all.

I mean, it's like, but then when we look at the stats is that developers are spending less than a third of the time writing the code, which is the thing that they love the most. And it's precisely because of that, because they are, I mean, either writing, volley, play code, writing documentation, finding bug, and I feel that ai, well, I find AI is helping with that. It is optimizing the time that developers are spending on doing the things that they like the most.

It's not just the coding, but it's the problem. Solving, coding for the sake of coding is not fun, but it's like when you apply your skills to understand the problem and solve that problem with code, that's when magic happens. Mm-hmm.

What do we need to do to our core CICD pipelines to enable all this? 'cause there's one way recently told me, he said, basically, yeah, we're running more code than ever. So we're basically running into the same wall faster.

Yes. But is, uh, the same way, we're also using ai, not just for writing code faster, we're using AI to optimize your pipelines. We are using AI to validate the security or compliance of your software.

So embracing AI is not just, uh, as part of your coding, um, coding, coding part of the job. It's embracing AI across your entire software development life cycle. Uh, even if you're able to produce much, I mean code much faster.

And, um, the problem to your point before is that there's always a bottleneck. Is that how fast you can deploy that software? How fast can you, um, secure your software?

How fast can you, um, fix a bug when that happens in production? And that's why it's critical not just to embrace or use AI in one, you know, in one of the, in one of the stages of your software development life cycle, you have to be strategic and mindful about using it across your entire software development life cycle. 'cause otherwise, your productivity gains on one side will become, you know, livity losses on the, on the front end.

Will we need to revisit the pipelines themselves? Because, let's be honest, um, a lot of them today are fairly brittle and they kind of break often and turns out we have more, probably more of them than we actually need. So do we need maybe longer term fewer, better pipelines or just more pipelines?

I think that's the, the, the, the beauty of using ai like the same with code. Do we need more lines of code or fewer lines of code, or we need, uh, with one framework or another? I think every application and every environment is different.

I think AI will help you, uh, analyze what you have and optimize that challenge that, that we have with pipelines in many cases, is that they got so huge that, uh, some people, even the people who wrote them, don't even understand them anymore. They, with the help of ai, you will be able to truly understand what's going on, if there's a problem, fix it without you having to go through your entire pipeline definition to understand what happened and eventually optimize them. Mm-hmm.

What's your take on vibe coding? Where does that fit in, in the DevOps workflow? Because in one sense, it's a tool for citizen developers and they might be creating code, but not sure about the quality of that code.

And does that mean that all those vibe coding tools are now essentially pushing code into some sort of C-I-S-C-D pipeline that another set of software engineers and their agent buddies are gonna have to review? Uh, um, well, there's, there's like, like everything else, right? It's, there's positives and negatives with that.

I feel that by coding is also helping accelerate the idea generation. We've seen companies that used to write every single spec, and now it is much easier to explain what you want to develop just by, by coding it so people understand truly what you, you, what you meant instead of writing an spec. And the other thing is that the problem is people may believe that those tools, we create a code by themselves, and then it just, you know, things are gonna just work.

And we know that server development is more than, than just that, especially when it comes to enterprise over development. And yeah, I think they are good in terms of helping developers initiate the process, but those tools are also becoming more and more complex and more and more professional. And you can also inject, um, custom rules, um, architectural guidance.

So the vibe coding experience is going to be, is going to get better and better and better. Yeah. Some folks will say that we're about to build and deploy more software in the next two years than we might have in the last decade.

Um, that's an optimistic view of things. But, um, I, will we get to that point? Or is it gonna take a little longer?

Because we gotta optimize the pipelines? But you know, it, it's like most things, it's a journey and we always underestimate the amount of time it takes to get to one place, to another. Um, it's a journey, right?

And, uh, as I say, it's a marathon, not a sprint. I see what it's happening today is that companies are able to reduce technical debt significantly. I think they're able to innovate much faster.

I don't know whether that's gonna translate into, uh, much, um, much, much software than, than than we had before. I think what we're gonna see is software being, uh, adapted much faster than we saw in the past. Mm-hmm.

And to that point, are we gonna focus more of our time and energy on new applications to replace the, probably many of our existing legacy applications, orations, will we go in and kind of modernize these legacy applications? 'cause the time and effort to do that might not be as intense as it once was. I think it's gonna be a little bit of both.

I happen to be with a global system indicator the other day, and one of the challenges that they face when they speak with customers is that they're dealing with massive applications, hundreds of lines of code that they can't touch because the people who develop those applications are no longer in the picture. They retired a while back ago, and now what they're facing is that, well, we need to modernize those applications in order to innovate instead of replacing all that code, which is almost impossible. So I feel that we're gonna see a bit of both.

We're gonna start seeing more applications being modernized, and once they get to that point, people will keep working on those applications, make them better and better. Yeah. And then the one thing everybody seems to be trying to figure out is there's no shortage of new vendors showing up with AI coding tools and platforms and these things.

And then there's the traditional existing DevOps guard, um, who's gonna win this fight ultimately, because, you know, on one hand the upstarts have the quote unquote cool factor, but on the other hand, companies like yours have all the data, Right? And I think that is the critical part that the idea of having a single data store that will provide companies full visibility of your entire software development life cycle. I see a world where those two things will coexist.

There will be new tools that will help developers do things better and faster. But eventually, as I said before, software development is complex. It's not easy.

It's not an app that you develop, you know, to run on your desktop for yourself. When you're thinking about enterprise grade software applications. You're gonna need a platform that will give you the, the security and the consistency, compliance and that, that companies need.

And that's why I see a world where Glab will become the core platform that will enable those scenarios where developers will use different tools. AI providers will bring different models. Uh, agent developers will come with their agents, but you will need a true orchestrator that will bring all those capabilities together.

All right, folks, you heard it here. The one thing that is certain is DevOps and software engineering is gonna change exactly how and precisely when remains to be determined. Amelia, thanks for being on the show.

Mike, thank you very much for having me here again. It's been a pleasure. Thank you.

All Right. And back to you guys in the studio. Vast weaves a deal.

The vast crypto spectrum, HPE decides to go its own way. Red Hat makes some updates to OpenShift. AWS is under the sea.

Pentagon is farming out ai. And we're gonna take a closer look at some of the announcements from KubeCon 2025 in this episode of the Tech Field Day Rundown. Hello everyone, and welcome to the Tech Field Day rundown.

It is November the 12th, and we hope that you're having a very happy lunch hour or whatever hour it happens to be, because it is national happy Hour day. Who knew that, that that was a thing? We're actually having a very happy hour here at Commvault Shift in New York City.

Uh, it's, it's a lovely blustery day, uh, just basically like it always is. And joining me, of course, is my co-host for this episode. Steven, it's good to see you again.

It Is nice to be here. I haven't been on the rundown in a while. Uh, those of you who, um, follow such things, you can mark that down on my scorecard, uh, because Tom and Al have been handling it so capably, but it's really nice to be back here for the rundown.

It is because we've got a lot of news and some of these things are definitely right up Steven's Alley. 17 billion. It is a major step forward in AI infrastructure.

The multi-year deal makes Core Weave Vasts largest customer, and expands the use of Vasts AI operating system, which unifies storage, caching, and data services to support large scale GPU clusters. As AI workloads grow, this agreement shows that fast reliable storage is becoming just as important as any of the compute power experts now expect storage to make up between three and 5% of total AI infrastructure spending, which is signaling its importance in the Gen AI era. Steven, you are intimately familiar with Vast Data, and as AI continues to grow, how does this deal help Vast Data position itself in the industry?

Well, uh, obviously, uh, by having vast, uh, sign such a major customer, such a critical one for ai, it gives them a real leg up in terms not just of, uh, being part of the AI infrastructure space, but in sort of cementing their future in that space as well, because essentially they instantly become one of the most important, uh, components of the stack. I expect that this will be a, uh, this will continue to be a, an accelerator for Vast Data's growth. And, um, frankly, uh, just on the face of it, it's nice to have such a big customer.

Uh, the flip side of, I suppose, is that it makes vast even more reliant on the AI infrastructure world and on Core Weave specifically. Uh, but I guess there's worse, uh, partners to connect yourself to. And, uh, frankly, if the, uh, checks are cash and the money's flowing, uh, sounds like a good move for them.

One of the things that caught my eye of this announcement though, is that it's not just vast supplying, uh, storage to support Core Weave, it's that they're gonna be using some of the special features of the Vast platform for this. Now, way back when Vast announced that they were gonna be more than storage, and they said that they were gonna have data, they were going to have, um, a, uh, event broker that they could use to, uh, support, uh, applications with structured data. Uh, they had Kafka, native Kafka inter interface.

My question, uh, for Vast at the time was essentially, that's cool technology, but is somebody really gonna use this? And is it really that much better than just using Native? And the answer is absolutely yes.

In fact, uh, vast claims that they're about 10 times better at handling Kafka requests on the same hardware as just running Kafka na natively, which is really a major advantage. And again, the importance of this shouldn't be understated. It shows that vasts pivot away from traditional storage and into data applications.

And AI integration is not just functional, but it's really beneficial to customers. And so, for that reason, I think this, um, says more actually in terms of the benefit to Vast than you might think. Um, you know, even beyond the fact that they now have a billion dollar customer, Tom Spectrum Labs Spectrum Fusion platform lets organizations verify their cyber resilience using cryptographic proofs.

This system tracks key tasks like backup and recovery, while AI agents monitor workflows and provide real time updates, the approach helps it teams demonstrate readiness, reassure boards and insurers, and improve recovery from cyber attacks. What's your take on Spectrum Fusion? If you ask me what the most important thing about the blockchain, what its legacy is gonna be, it has absolutely nothing to do with any magical pretend money.

And it has everything to do with the immutability of the ledger, because one of the things that we need more than anything else in it is the ability to say for a fact, the thing did or didn't happen. And what this platform is allowing you to do is create a token that you can then share with auditors and insurers that say, for a given point in time, we met the requirements that you put in place for this. Now, why would that be important?

And especially when you start talking about things like cyber insurance, well, I don't know if you're the underwriter for one of these policies, you're gonna wanna make sure that the company that you are underwriting did everything they could to prevent whatever it was from happening, from happening. And this token kind of serves that purpose, if you will. So if I am a insurer, I would say, I would want you to use this platform.

What you do is you go ahead and install it. You run it, and it gives you the token. And the token says that, you know, on, I don't know, November the 12th, 2025, you met all of the requirements to be insured with our policy.

And then two weeks from now, someone comes out with this massive zero day that, that nobody could have foreseen and violates all of your security policies. Well, as an insurer, I know that at that point in time, you met what I needed for this. That's an easy, uh, claim to, to file, right?

In fact, that's one of the things that their AI agents will allow you to do is file that claim automatically. Uh, if you've ever worked in insurance or you've ever had to deal with insurance, you know how big of a deal that can be. But it also allows you to define what things you are going to be, uh, running these token proofs against.

It's not just, do we have a firewall? It's not just do we have separation of policy? It is, are the backups working?

Have they been tested? All of these other things that I would want my system to be able to prove beyond a shadow of a doubt that you're capable of doing, because that way everybody knows, and because they can be shared, because they can be rerun at any time, they help provide that proof. And this just leads to more resilient platforms.

That's really what we want. Insurance is there when the resilience fails, but if we can prove that it was resilient at some point in the past, then we know the gaps that we need to address and cover so that everything works out well. So I, I'm kind of happy to see this.

I'm also kind of happy to see a, uh, solution that involves using cryptography that isn't about Monopoly money. Steven, we had news this week that HP is gonna stop selling Qumulo ity and WCA software by March of 2026 to focus on its own storage pro platforms like Electra store. Once Zerto and GreenLake, the change simplifies HP's portfolio and strengthens its in-house offerings.

Scalability and CCA will continue working with HPE through channel partnerships, though the move might impact their sales. And we know that HPE has been working really hard lately to kind of simplify their offerings because they do offer quite a few things to their customers. What does this move signal for the industry?

Well, um, maybe a little less than you might think. Um, fundamentally, HPE has always been a good partner for many companies in the industry, including, of course, companies like Scale, acal, Culo, and cca. Um, in fact, I believe that, uh, scalability and HPE, uh, really sort of blazed a trail there.

I I think that HPE must have been S scale's first major, uh, OEM, uh, their major channel and, and frankly helped build that company. Uh, similarly, you know, CCA has had just tremendous success in the AI space. Uh, you know, Qumulo has a long, long history of developing great products as well.

And of course, HPE has an OEM version of Vast Data, as well as that they sell in addition to their own, uh, Electra storage platform. Uh, this change essentially is, uh, HPE saying, you know what? We should really be focusing on our internal, um, products and, and a little less on some of these partner products.

But that doesn't mean that it's a fundamental change. In fact, I suspect that Channel Partners, I mean, HPE obviously is one of the biggest channel companies out there. I think that their channel partners are gonna continue selling wca Ity, Qumulo and more on HPE hardware going forward.

So I, I really wouldn't read too much into it. Instead, what I think this is, is it is just sort of a cleanup of the way that HPE goes to market, the way that HT HPE works with, uh, both partners and OEMs and the channel, and shows that HPE is increasingly confident in their own internal storage platforms. So overall, um, I don't think this is gonna have a huge impact on customers.

I also don't think that it's really gonna have a huge impact on the companies that were named here that are, uh, allegedly losing access to HPE customers because it's not like those customers are gonna switch to something else. They're probably going to be working with HPE and, um, you know, WCA or Ity or Qumulo for the long run. Um, I, I think the one interesting aspect here is that HPE apparently is going to continue working, uh, with their OEM version of Vast.

Now, I don't want this episode to sound too much like a, a cheerleading session for Vast, but boy, that does, uh, lead some credence to the fact that, uh, vast is sort of the golden child of the storage industry right now when HPE is continuing to work with them, even as they are separating themselves somewhat from some of these more traditional partners. 20 adds new AI tools, stronger security, and better Edge support. Key features include AI workload support, post Quantum Cryptography, zero Trust, identity, and Easier Secrets Management.

This update also expands ARM and Oracle Cloud Support BGP networking and two Node Edge deployments, helping organizations modernize applications and manage growing Kubernetes environments more efficiently. Uh, what's your reaction to Red Hat's OpenShift announcements and the overall strategy that Red Hat has for OpenShift? I think it's important that this came out when it did, because it signals that Red Hat is not giving up on OpenShift.

I know I was looking at my travel logs from, uh, years past, and it was only nine years ago that Stephen and I were at, uh, the, uh, OpenShift Summit, uh, OpenStax Summit down in Austin. And of course, OpenShift is, is definitely related to all of that. But I think that this is, if Red Hat's gonna offer cloud in a box to people, it has to be a modern cloud in a box, right?

You have to be willing to take on all the things that are available in, uh, Azure, in AWS, in GCE in any of those platforms. And the fact that they're supporting Oracle Cloud will tell you that, oh, it looks like Oracle Cloud's actually coming along with all these deals that they've been making. So looking at the features like post quantum cryptography is kind of table stakes right now.

We've seen this shift with a lot of people where they realize that the time to get in front of this problem is right now, because the implementation of, of post quantum algorithms is not gonna cost you anything right now. It will cost you a lot whenever those data, uh, storage units are encrypted with older technology, and you have to burn CPU cycles to get them upgraded. Now's the time to do that.

Of course, you know, you wanna talk about things like zero trust secrets management. Yeah. That is, again, table stakes, right?

You, you have to have some kind of an isolation mechanism for these workloads. You have to be able to do secrets management, as we've learned from all of our friends that have done a number of presentations at Security Field Day. Uh, like one password, for example.

If you can't do Secrets management, nobody's gonna rely on you. Again, adding arm support, adding BGP support, adding two node edge deployments, these are the things that I would expect from a modern platform. And I know that the people at OpenShift have been working really, really hard to kind of be the alternative to doing things in the public cloud.

They're your private on-premises cloud. And, and this gives me hope that this is not something that's just gonna disappear. I obviously, with Red Hat and IBM behind it, it won't, but this can be positioned as an alternative for people who are under massive regulatory strain, that that cannot be in the public cloud or can't have the majority of their data.

And that's especially true for people who are sensitive about having their, uh, data uploaded to AI clusters everywhere. So, good news from Red Hat, if you're an OpenShift customer, I'm sure you're gonna wanna upgrade pretty soon because there's a lot of new bells and whistles that you're gonna wanna try out. The Pentagon is pushing AI adoption in the military, but they're expecting companies to pay for the development and train that AI themselves.

According to Defense Secretary Pete, he tech and defense firms, including Google, AWS Meta, Palantir, and OpenAI are going to have to provide these contracts to the Department of Defense. But the basic training needs need to be done by the companies themselves. The focus, of course, is on fast acquisition using commercial technologies and deploying AI in things like drones, autonomous vessels and other systems.

The companies are supporting the effort, but experts are warning that they're gonna have to balance speed with quality and of course, shareholder expectations. Steven, do you think that the Defense Department is on the right track here, making the companies do all that training themselves? Or should the Defense Department pony up if they expect the models to be trained?

Well, I think that, uh, this is a little scary to think about. Um, you know, you, you don't necessarily want to chat GPT deciding whether to shoot you or not. Um, or at least I wouldn't want that.

Uh, but on the flip side, I think that this is more about trying to figure out how we can leverage, uh, commercial technologies, as you said, for military applications and less about, uh, giving Claude a gun. Um, you know, that being said, uh, this is fairly typical for, um, government to work with the private sector in various different ways. Uh, typically when the government has a, uh, job that needs to be done that's specialized that the private sector can't have anticipated, they do, uh, support that in many ways, including paying for development, uh, or building in, uh, you know, building it on a cost plus basis so that the private sector can meet the needs of the military that are usually dramatically different than civilians.

Uh, I think that ultimately what we're gonna see here is, um, applications of the a of AI technology that are very similar to what you would find in the civilian space. Um, this is not going to necessarily be, um, you know, using, uh, chatbot to do some things that you would maybe not, it would not be prepared for. But that being said, there are a lot of companies out there that are actively working on developing military applications.

Uh, you know, we've of course talked about, you know, uh, the big guys, you know, anthropic and AWS and Google Cloud Meta, Microsoft, Oracle. Um, Palantir is one company that has put in a lot of their own development resources to support the needs of government and the military. And of course, there are others, um, specialized companies that are developing drone technologies.

I was, uh, talking to somebody who works with one of them, um, in Ukraine, uh, recently. And, uh, they are doing just remarkable things with AI and drones, um, terrifying but remarkable things with the drones in Ukraine, uh, using AI technology. And I imagine that that's pretty attractive to the US government as well.

So really, I think that what this means is, uh, the US wants to be able to leverage AI as quickly as possible with in as many places as possible. And hopefully they will approach it in a way that is, uh, at least technically going to result in, uh, a, a usable and not, um, Skynet type product. Tom, uh, Amazon Web Services has announced fastnet.

It's first fully owned transatlantic sub C cable. It links, uh, Maryland to Cork County Ireland. It's set to go live in 2028, and it will deliver 320 terabytes per second of bandwidth, which is enough to stream 12 million HD movies, which apparently is our metric for bandwidth.

Now. Uh, this strengthens AWS's global network and supports growing cloud and AI workloads. Uh, the 4,000 mile cable adds redundancy as well, uh, outside of traditional routes, improving resilience against outages or damage.

It's built with advanced optical switching technology, and it reflects, uh, AWS's push to get more control over their own in infrastructure in-house. Uh, what's your reaction to fast net? Sounds like a fast net to me.

It does, and this is a big deal for Amazon because this is their cable. They are not renting it from anybody else. They own this thing from Maryland to the island of Fastnet where there is a lighthouse and assuming some other things that are kind of awesome.

Uh, I don't know if they're gonna put the data center in the lighthouse, though. It's Kind of a cool coincidence that the island is called Fastnet, Isn't it? Like, it's almost like they did that on purpose.

Uh, here's the thing that I worry about with Amazon, though. Uh, one of the reasons why you don't see a whole lot of privately owned under sea cables is because they're hideously expensive. Uh, I did not notice in the article did Jeff Bezos get a Navy, because he's gonna need one in order to lay this cable.

It's 4,000 miles long. That is several meters from my back of the envelope math. But what's really important is the fact that it's gonna be armored, right?

Like you have to armor this thing to avoid, um, rogue navies from dragging their anchors on the sea floor and taking it out, or, or who knows what else. But that is not a cheap process, and you're gonna have to lay all that cable. I'm sure if they started now, they'll probably do be done in like three years when it opens up, Amazon's gonna have all of this massive bandwidth to still make people use US East One as the default instance.

Uh, I, I think what they need to do though, is they need to start selling this as you can start hosting these, uh, you know, production units and everything in Ireland or in European data centers. And, you know, you can move data back and forth as quickly as possible. Yeah, 12 million HD movies sounds like a lot, but how many AI workloads can you move?

Because one of the things that's cited in the article is the fact that AI workloads can get awfully bursty whenever they find a new source of data and bandwidth to, to jump into. And they're gonna have to do some really impressive things as far as like kind of metering all of that. I'm curious to see how, if this is an investment that pays off, of course, if I had Jeff Bezos's money, uh, I have to do something other than launching celebrity singers into space on rockets.

Um, this is probably something a little more pragmatic that Andy Chassis is really gonna appreciate, because it's gonna increase AWS's bottom line overall. All right, we, uh, had a couple stories we wanted to take a closer look at, because this week is KubeCon, it's in Atlanta, so we hope that you're all enjoying some peaches and, uh, Braves baseball. Uh, but there's been a lot of announcements that have come out of, uh, KubeCon that we're definitely gonna wanna check out.

The first is that the Cloud native computing foundation, CNCF has introduced to certified Kubernetes AI conformance program to standardize how AI and ML workloads are gonna be running on Kubernetes clusters. The program ensures workloads are portable across different environments, including hybrid and sovereign clouds, backed by companies like everybody's favorites, Google, Microsoft, Oracle, Broadcom, and Red Hat. It aims to prevent platform lock-in and simplify large scale deployments with AI workloads increasingly managed by centralized IT teams.

The initiative supports the magic word, interoperability, scalability, and efficiency through deployment across Kubernetes ecosystems. Steven, do you think that the CNCF has enough clout to standardize how we deploy AI on Kubernetes? Well, I think they have enough clout to standardize how we deploy on AI on Kubernetes.

I am not sure they have enough clout to standardize how we deploy ai. And I think that that's an interesting and important distinction to make. Essentially, uh, AI applications often run on, uh, dedicated platforms that, uh, have no Kubernetes in sight.

Some of them do, some of them don't. Uh, when they do have Kubernetes, uh, many of them have that somewhat hidden from the implementation, uh, from the, the, the higher level details. Some of them, uh, allow the customers to see the fact that it's, uh, that, that all that's running.

Um, I think that ultimately what this is, is it's a reaction by the CNCF to the fact that one of the most important, um, new workloads is not necessarily Kubernetes native. And the companies that you mentioned, uh, all have bet big on Kubernetes as the future of, uh, infrastructure. Uh, and so, uh, they wanna make sure that they're ready for that when it comes to AI applications.

So, uh, if you were cynical and if CNCF was a corporate entity, you might say, well, this is just a, a ploy for them to maintain relevance. But CNCF is not a corporate entity. In fact, CNCF is a pretty remarkable industry collaboration.

And I feel that in fact, this is not a cynical approach. I feel that this is, uh, an example of, uh, need within the industry for a, uh, portable way to run AI workloads. It's a need or a desire within the industry to have AI and cloud share similar infrastructure.

And ultimately, I think it's good for the industry. And it's the sort of thing that CNCF ought to be doing. Hi, it's me, cec, but not for, for the reason that you're thinking, I'm actually cynical in the opposite direction because I love that CNCF has decided to do this.

They're like, we are essentially the stewards of Kubernetes. We need to keep Kubernetes relevant because, well, it's basically the cloud right now. What if you're running something in the cloud, it's probably running on Kubernetes.

The problem is that they see where the future is headed and the future is crashing through the streets like a bull in a China shop, because AI doesn't care. It doesn't care about frameworks, it doesn't care about platforms. All it cares about is consume more, more power, more ram more storage.

And that doesn't play well with things that have frameworks around them because there's a way that we do stuff. And I think that C Ncfs basically saying, if you're gonna use Kubernetes to do ai, you've gotta make it play well with everything else so that it doesn't just basically act like a swarm of locusts and consume everything in sight. I wonder though, if the companies that they are targeting with this, everybody's favorites, Amazon, Microsoft, Google, Oracle, are gonna play well with that, knowing that if they make CNCF mad, they might lose future input into the way that Kubernetes works.

Because I got news for you folks. AI isn't the only workload out there. There are a lot of things that are still gonna be running on containers long after whatever the new hot thing in AI is, is put out to pasture.

So if I were the companies and, and I know you're watching this, uh, Larry Ellison, uh, char Andy Jassy, all of my friends don't make CNCF mad play. Nice. You can still have access to all the ram you can get your hands on.

Just, uh, just go with it a little bit. Here's another announcement from KubeCon 2025, and it's from our friends over at ARM because they highlighted their Neo verse platform and Google Cloud's new Axion in four A VMs, which are enabling energy efficient, scalable AI and cloud workloads on wait for IT. Kubernetes collaborations with CNCF partners like Harbor, OPA, Edify and AuthZ help developers build secure, portable cost effective cloud native systems from edge to cloud.

The focus is on smarter, more sustainable architectures for modern applications. So we just talked about CNCF putting all this framework in place for Kubernetes, and now we have ARM saying, guess what folks? We're ready to adopt it.

Think this is a good move for arm. Oh, yeah. Um, yeah.

ARM has obviously, uh, come to dominate mobile computing. Uh, they are taking a serious stab at dominating, uh, desktop and client computing as well. Uh, they're an awful lot of ARM devices in the Edge network, and frankly, they are heading to the data center and AI as well with this, with this platform.

Um, ARM is very, very smart to recognize that the thing that keeps people from selecting different hardware platforms is often not hardware, but software. I mean, the hardware has to be good enough, but the, uh, software also has to meet the needs of, uh, developers and of end users. And that's exactly what ARM is doing here.

Essentially, they are working with the industry in a productive way to make sure that everybody can use their platform and that leads people naturally to want to use that platform. Now, another aspect of this, of course, is the announcement that by Google that they're gonna be having their new Axion N four a, uh, cloud VMs, um, built on the neo averse platform. Uh, we have seen, um, the proliferation of ARM-based instances in the cloud, but generally those haven't had the impact that we might have thought they would.

I think one of the reasons for that is because customers honestly don't really care about performance per watt. They just care about performance per dollar. And to this point, that has been a bit of a stumbling point.

It's sort of blunted the impact of arm cores in the cloud, but if ARM can do more than, uh, the competing X 86 based platforms in areas like AI inferencing or, um, you know, other areas, if they can deliver the goods at less money, I think that that could be something that could be exciting to customers, especially if it just works. And that's the challenge. I think it is too.

And I love that ARM is embracing this because, like you said, the optics around, uh, reduced cost per watt, per whatever don't really matter. The electric bill is what matters. And the electric bill is a component of how much I'm gonna rent these things out to it.

So if ARM really wants to make an impact in this, they have gotta figure out a way to cut this cost to the bone to make these things way more appealing to companies that are saying, well, where should I put these workloads? 'cause if it's all running on Kubernetes, honestly, nobody cares, right? We're gonna compile it on whatever platform we need to use it for this week, but if you can consistently deliver the things that make it a lot cheaper, then you are gonna want to embrace those platforms, right?

We saw this originally with X 86. It won because it was more efficient and cheaper than all the other options that were out there. And then it became the lumbering behemoth that it became.

And now ARM is kind of sneaking in to take some of that steam. And if you don't believe me, look around your desk and realize that with maybe one exception, everything you are running is running on an arm core, whether it's your mobile device, your tablet, your laptop, and even if you have one of those X 86 laptops that's running a Redmond operating system, there's probably an arm core in it that's doing a lot of AI offload. So ARM is kind of the way that we're looking at this.

And I quote, everyone's favorite, 1995 seminal movie Hackers. Risk Architecture really did change everything, and it appears to be poised to take over the world. Uh, congratulations an Angelina Jolie and Johnny Lee Miller, you you live in our hearts forever.

One more thing I wanna mention here at the end, Tom is, uh, we are at CubeCon with, uh, tech Field Day this week. Uh, that's one reason that, uh, we are, uh, recording these things at a different time of day. But the other reason is because you and I are not at CubeCon.

We are here at Commvault Shift in New York City this week. Uh, Commvault's announcements unfortunately are embargoed past the date that we are going to be releasing this episode. So I want to give a shout out to Commvault thank you for hosting us, and we, I promise we will cover your stuff next week.

Um, next week is also the day, you know, on, on Wednesday that, uh, Commvault's going to be live streaming their, um, or not live streaming, but streaming their, uh, announcements from shift. And we're gonna be doing a live blog of those announcements on the Tech Field Day website. So, uh, we are very excited to be here at Commvault Shift this week, but, uh, we are not covering it on the rundown this week, but we will cover it next week.

We take embargo seriously. Mm-hmm. We, on the other hand, have had a number of other Tech Field Day presentations from, uh, CubeCon this week.

Again, uh, the timing of the recording and production of this means that we weren't able to cover those on the episode this week, but I'm sure that we will be talking about some of those things next week as well. So, just a brief program note, if, if you're wondering why it is that, uh, we're not talking about Convault Shift and, uh, why we didn't cover the Tech Field Day at CubeCon, that's why. Yeah.

And make sure you tune in for that live blog on the 19th, because I'll be sharing my thoughts about all this cool stuff that I know about that I can't tell you about yet. But trust me, you're gonna wanna tune in. com for a lineup of all the events that we have coming up, because guess what, folks, we're getting close to the end of the year.

You know what that means? We're already planning for 2026. We have a lot of things in motion.

We have a lot of events that are already scheduled. com, you can see the list of those. I've got some stuff going on.

Steven has some stuff going on. Our good friend from down under Alistair Cook has some stuff going on. Find the event that works the best for you and put it on your calendar so you don't miss out on that opportunity.

Just like we know that you have a reminder to tune in to watch the Tech Field Day rundown every Wednesday, we love making these episodes for you. We love finding all the cool news that we wanna share with you, and then we post it to YouTube, uh, or, you know, maybe your favorite podcast application of choice. Uh, there's a lot of input that's been going into Apple Podcast, overcast recently, and we love to hear all about it.

But we want you to tell everybody else what we do here. That means leaving us a rating and a review, all the stars, all of the thumbs up, tell people how cool we are, because if you do that, then we get put in front of more people who love to hear about the ins and outs of enterprise tech news. Um, that's, uh, an underserved market.

And, and I think we're doing most of the serving there. Uh, don't forget though, that Stephen and I also have a lot of other things going on with, uh, other Futurum group properties. I have a podcast that's related to Security, security Boulevard.

We just released a new episode with my friend, ed Whedon. You are a regular, uh, guest on Textron Gang, along with my good friend Alan, who's a co-host on Security Boulevard. And We just launched an AI podcast with fu with, uh, Futurum, uh, and Textron as well.

So look for utilizing AI in your favorite podcast application. It's me, uh, Nick, patience, Olivier Blanchard, um, Mike Fazar, et cetera, talking about, uh, AI in a similar vein to your Security Boulevard podcast. Absolutely.

We're gonna be back next Wednesday with all the news that was in the it week that we just finished up. And we'll be talking about Commvault Shift as well. Uh, make sure that you tune in.

But for myself, for Steven Foskett, for Matt Garvin, who's been hanging out in the back, making sure all the cameras are straight and all the audio levels are recording properly. Thank you very much and we'll see you next week. He's really Here.

Stop poking. He's really here. We're not, we're in the same room.

Hey everyone. Remember when tech used to fight the good fight? They were the good guys.

You're watching Textron Gary. Hi everyone. Happy Monday.

It's Alan Shimel. Guys, I am glad to be home if only for a day. Uh, it was, it was cold in Atlanta and I, you know, it was also really bad wifi at that conference.

It was, it was hard. It was hard maneuvering and working, but it was a great coup con. I've written about it.

We spoke about it all last week on our gangs. You could check it out. Um, but we're back here in studio and we've got a great lineup of our gang members to talk about some good stuff.

Let me introduce you. We have a, a DevOps dozen finalist, Garima Boal, a security analyst, Jack Poller, an analyst who's finding her next career, or her next joy in life, our friend Kimberly Bates. Would that make you like a recovering analyst, Kimberly?

Yeah. Yeah. A recovering analyst.

There's hope. That's good to know. Yeah.

And then, um, and then two more analysts Were always with us. The one and only Mitch Ashley and Guy Coer. Guys, how are you?

Good, good. So, I don't know, maybe it was the kumbaya of the open source movement and the idea of doing things for the common good that got me, or maybe I'm just tired of seeing the pig's feet at the T trial, but I, I, I did a, a shimmy says, uh, a i it article and a LinkedIn article last week on this idea of, you know, tech used to be the good guys, even in the movies, right? You go to the Avengers, I'm Iron Man, right?

He was a good guy. He was fighting crime, fighting the good fight. He, to me, that's what Elon Musk should be aspiring to be, right?

Or, or any of these tech mogul, tech bro billionaires that we have kind of in, in place today. But that's not the tech we get today, right? Increasingly, you know, it's a, it's a selfish kind of egotistical egomaniac.

Egomaniacal, if you will, kind of driven industry. We don't have any numbers because the government's not releasing any numbers, but stories, you know, garnered for from like employer records. And, you know, some of the big PEOs and stuff say that we might have lost maybe 35,000 more tech jobs last month.

Last month alone. Things aren't so great in tech, in spite of the fact that we're spending record money building data centers and this AI thing. And I, you know, increasingly people are recognizing, and we'll talk about it later, you know, we might be in an AI bubble, who knows, but Mitch, guy, Kimberly, Jack Garima, we've all been in tech collectively.

There's a hundred years, maybe 200 years worth of tech experience here. Have we, have we, are we not the good guys anymore, or were we ever, maybe we were never the good guys, But first came to mind. Um, you know, we've all had people that we've admired over the years, whether you're like Steve Jobs or whoever it might be, right?

Alan Kay as a researcher. Um, I think we're, we're in the Gordon Gecko phase of tech leaders. Like, where's my fricking trillion dollars?

He is Good. Yeah, exactly. For my bonus.

And, you know, I, I think for a while, Elon Musk was held up as, you know, someone that's admirable because of cool things that he was doing with space and, and, uh, you know, Tesla, things like that. Now he's just kind of another one of the greedy people on, on the block and, uh, doing, doing side deals with everybody and, you know, forcing what he wants out of his board, just to pick on one. But I, I don't know if we were ever really, truly, you know, tech were the good guys necessarily.

I'm not saying we're the bad guys, but I think we had people that we admired and we held up as whether role models or not, I don't know, but people that we just admired a great deal because of what, what they did for the industry and their leadership in the industry. That's what I think we're missing. So I'm, I'm gonna be the contrary here a little bit, because the first thing that came to mind is that when I joined IBM way back when, and I represented at least a quarter of the years that you talked about here, No, you don't.

We're all your age though, close To it. Close To a quarter. The three of us make up three.

We were on our antitrust investigation, and they were about to get split up into three. And there was a reason why they were doing that is because IBM was not playing nice. We were wrapping up the things, the big, you know, the burrows, the Honeywells, that kind of stuff.

We were screaming over them. I mean, and then I started thinking about, you know, Tom Watson, he had his entire desk dropped in the front yard at NCR when he worked for them. That was not exactly very nice either.

And then you add the, okay, okay, 1998, I was looking this up, Microsoft versus a Mosaic. Netscape, you had 20 states going after Bill Gates. Bill Gates is saying crush 'em, just crush him, you know, he was really nice back there.

Now he's like a really good, nice little woke guy and was like, well, baloney. Sorry, I almost went the other way. And then, then I'm gonna add one more to my favorites, is Uncle Larry take no prisoner, Mr.

Larry Ellison kind of guy. I mean, he's been dropping off people every quarter. Think It's Grandpa Larry by now, He's Grandpa Larry, sorry, uncle Larry, whatever.

No, that it's media mogul, Larry Media, mogul, Leary Leary, whatever. I mean, every quarter you wipe, he wipes out a, you know, a percentage of his people there. And that's been going on forever.

And he, he just never has to publish it because he does it every quarter the same way that, you know, uh, the guy at GE did it. You know, it's the same kind of principle. So, and like, you know, I think what happened is that we went, and I'll shut up after this.

We went through this period where everybody was like, let's be nice. Let's be sweet, let's be okay. And now we're back into let's go compete.

And, um, that's, that's where we're at. And you know, I embrace it. I embrace the competition.

And, um, that's, Yeah, I don't, you know, I, I've only, let's, let's let everyone else trophy before I jump back in, but thoughts from the rest of the team. I'll, I, I, I will say that I'm on Kimberly's side and, and having been both at IBM around the same times, and then at, uh, Netscape during that time, and, uh, my new employee orientation at Netscape was a speech by, uh, our now venerable vc, mark Andreessen, who said, uh, Netscape earns $70 million a year because from ad sales on our homepage, because nobody ever changes the homepage. And so we're suing Microsoft outta the browser to protect that revenue.

That was all it was about, right? Is protecting that revenue. It had nothing else to do with anything else.

And they knew they sued Microsoft just because they could. And when we, I did an IETF meeting where we were discussing, uh, um, uh, TLS, the TLS specs, and we had two coalitions, the pro Microsoft Coalition and the anti Microsoft Coalition. And we spent most of the time figuring out what Microsoft wanted and trying to design the protocol to make it more difficult for Microsoft.

Well, when Windows ain't, and that was until low level Engineering level, I'm sorry, windows ain't done till, notes don't run, right? That was the old thing at Microsoft back then with notes, because the 1, 2, 3 was killing Excel. But, but guys, you, you're missing, I, I think you're missing my point with all due respect, you know, to quote the infamous Don Barini from Godfather, after all, we are not communists, right?

There's nothing a matter with making a profit. There's nothing a matter with competing. Good old American competition may the best person and company and product win.

There's nothing a matter with using the law to your advantage. This is how the rules of the game are played. This is, and there's, and that's great.

That's America, that's capitalism. That's market forces at work. Don't fool yourselves into thinking that's what's going on here.

What we have here is what we call collusion. What we have here is Uncle Larry is in bed with Young Sam and Young Sam, as soon as he gets outta bed with Uncle Larry runs over to, to, whether it's the Microsoft or Google guy, and does his little jig there, and they pass money around this virtuous circle of the eight or nine of them creating barriers to entry for the two guys for the next laws and jobs sitting in a garage working on something or, or, or anyone else who wants to come in here. Right?

So how is that so much different than what was going on in Sand Hill with the VCs? I mean, that the VCs are just rolling doors of your, you know, it's, it kind of goes back to New York when you had the Morgans Morgans and the, you all, the, the people back in whatever years it was, and those were the people that socialize with each other. And that's the same way it is right now today, detective, But not at this level.

And then I'll tell you what else is the big difference. They've got a new partner in the circle who's taken a piece of the action, right? The federal government, the federal government used to, and the court system used to be the, the arbitrary, the, the, the, you know, the guys who made sure at least, even if it wasn't real, at least the facade was Horatio Alger work hard and you'll get ahead.

Now. They got their thumb on the scale too. I, I feel like the, the, the discussion here has shifted because my take on your take, Alan, My take on your take and Your take.

Go ahead. Yeah. Well, and, and how you summarize it here and in your shimmy says, and how you summarize it here was you were talking about, um, tech to benefit the world.

Tech used to be the good guys to benefit the world. Now it's the bad guys to benefit these nine people. That's what I thought you were talking About.

Yeah, no, it did. And and you are right, guy. We, I I, shame on me for not mentioning that too.

I think there was sort of this altruistic of we're changing the world with tech, right? Tech is the great kind of equalizer. We're going to get women and other underrepresented in groups, you know, working in it.

And, you know, and, and Kimberly look, I mean, no, I'm not saying anything bad or I don't hold it against me, but you came up in a time when it was really hard for a woman, too. No, you don't think it was hard for you. I, I am, I'm one of those contrary people on it.

I, I didn't hit the walls like I felt like, you know, other people talk about just either that or I just ignored it. It was Completely, well, God bless you and good for you, Clueless. I was probably just clueless to it.

It just kind of blinked, Oblivious to the Doy, just do your job. Gima what about you? Do you feel like, Yeah, I think, uh, you made the point very clear in the beginning that, you know, tech for good, tech for communities, and I represent community and community leadership here.

So I would say that, you know, with all this, what you wrote, what are the repercussions happening? It's fear, it's disengagement, it's normalizing corruption, uh, stifled creativity. And what happens when it goes to the peak communities, uh, you know, strike back and this is what will happen.

And I will give you an optimistic view on this, because I think you mentioned about open source movement. I come from that open source. I am, um, I'm the child for, from the open source movement, the communities.

And I feel that, you know, um, it's time for us to kind of ensure that something what happened in the 1980s with, you know, uh, Linux, uh, taking the power back, right? With community and open source, uh, trying to kind of, uh, uh, steer the needle in the right direction. I, I, I am more optimistic with leadership and the communities and history reminds us what happens, happened with DevOps, you know, all those tools, capabilities, this, uh, Kubernetes movement.

How did it got started? You know, it's massive right now. So I'm more optimistic.

It's just a matter of time. I mean, you cannot stifle innovation and adaptability on the name of egg diplomacy. And this, I mean, it's just, you know, how we see it today and what can happen when community comes back and try, tries, tries to take charge of it.

Amen. I I think I'm quite optimistic about this development that I think you pretty accurately described, Alan. Um, I kind of agree with everybody here, which is really weird.

Um, the, so I think the fundamental issue here, I think is that there's tech and there's science, and they are, they are very well linked together. The science brings the discoveries, like the original invention of the semiconductor that enables the tech, namely, you know, information technology scientists are pursuing knowledge, truth, betterment as a general rule. Yes, they have, you know, businesses as well, and they, they, they do commercial things.

But that's the general culture there. Um, we're, we're gonna see that this week. I'm attending Super Compute, which is one of the great fusions in the world of science and technology.

Technology or tech. That's a business like you described. And I think that we are reverting to the mean of understanding that the business is in the business of its own business and profit and science remains the pursuit of knowledge, fundamentally.

And all that happened was probably led by jobs, was this idea. I mean, Steve Jobs was this idea that, that, um, there's an altruism. I mean, that's one of the things I despise about the whole Silicon Valley culture is how this new platform, uh, and and Gizmo that I've invented is what's gonna bring peace to the Middle East and extend lifespans by 10 years and all this other kind of garbage that helps the VCs get all excited and the money flow, but it's really just garbage.

And so if we're gonna finally start recognizing them for what they're trying to do, which is to make themselves rich and get their trillion, then, you know, great. 'cause that's what they've been doing all along. They've just been papering it over with this idea that tech is also in the business of the common good, and it's the science that's in the business of the common good more than the tech.

But, but they, they, but as you started it off with, they're intrinsically linked, right? That's the tension. That's the tension.

So it's always been there, but you know, it's kind of like adopting this idea that this innovation, which is the word usually used, this new widget is, uh, like the, a scientific pursuit for the good of all. And it's not, it's not, it's the new widget. It benefits from the science.

We're seeing that in AI right now, You, you've hit it right on, which is that we in the Silicon Valley and the Silicon Valley tech leaders have put a fig leaf of altruistic respectability on top of what they do, because you, and there's a lot of reasons for that. But it's, this is history repeating itself over and over and over again. If you just look at the printing press, when the printing press was invented, it was of, I don't know if it was created for altruistic ideals, but it had a very, um, you know, it ushered in the age of enlightenment where, you know, in the education and the explosion of knowledge throughout the world.

But then if you look many years later, we had a closing down of information availability and our knowledge and media was controlled by three corporations in the United States for a very long time. And before that was controlled by William Randolph Hurst and the newspaper barons. And the explosion of the internet was what freed information out.

And now we can get news from a gazillion different channels, Better off we are, Yeah. But, but it's a cycle that repeats itself, right? So, you know, I I, I understand your frustration, Alan, but I think it's a little bit naive to think, oh my God, That this is different.

Maybe it's naive, right? Let me ask you a question. Do you think Hurst was a good guy for what he did?

None of these people are good Guys. Points of bad. They had none of, They had good, no, but None of our Features forever.

I mean, what, what are your opinion of the, let's beside the point, let me just tell you, hear me out their name by itself. Don't let it prejudice your thought, but the robber barons, they're called Rob. Do you think Getty was a good guy?

Do you think Getty was a good guy? How about Mellon? Yeah, those Melan was one of the Robert Barons.

I Mean, we have a reason why we have antitrust laws. Exactly. How about, it's the reason we used to have antitrust laws.

But, but, but then Again, and then after they make all this money, then they do this thing, this, you know, foundation and start giving away zillions of money Or, or build universities. My la mater was Carnegie Mell, excuse me, there you go. Carnegie Mellon University, right?

It was funded by the robber barons Or, or, or funded that building at the, at the university, university or whatever, put my name on it, right? It's like, Yeah, we've shifted from, we don't call him. I've seen it's the same thing, Robert Baron.

We call him Titans now, right? No, I still call him Robert Barons. Mitch.

No, we, yeah, some of it, I mean, I, I mean, that's Benioff and Mark Benioff, who's the billionaire who, uh, does Salesforce, right? Uh, he's single handedly funded a huge amount of the, the UCSF child children's hospitals. And, um, uh, Ken Langone who did, uh, home Depot is the Langone Cancer Center in New York, which is very good at treating cancer patients.

I mean, the, those billions of dollars that they've created have not gone entirely to gold toilets and lavish dinners. And, you know, Larry Ellison's yachts, yeah, he does have a yacht, but he also employed a couple thousand people, people to Build that yacht. After those, after the gold toilets and the lavish dinners and all that other sort of stuff, there's some money left over for cancer.

That's great. I'm not, I'm a capitalist. I'm not, I'm not trying to critique the entire system.

I, I just, I'm more focused on this idea that tech and the tech industry does do good. It should be, we, we benefit from it. But, um, expecting brands and companies, what I, what I like, what makes me optimistic is that I feel this cultural shift that you're detecting, Alan, where we stop believing when the next Google has their version of don't do evil, that we just stop believing that it's beside the point, it's marketing.

And if the marketing stops working, we don't have to put up with it anymore. Well, I think we could learn from mothers not in tech. Um, I'm gonna represent my home state.

I admire Oracle, not Larry Ellison, Oracle, but the Oracle of Omaha. Was he from western Nebraska? Western Omaha.

Okay. Yes. He's a little more Nebraska west of Western Omaha.

That was a joke that we had. No, We weren't talking about that in Cucu. It says he's from western Nebraska, Mean Warren, Warren Buffett.

You know, you talk about somebody who's made people a lot of money, including himself and and his investors. But you know, that's someone who, he, he had, he had a, uh, an attitude of not only philosophy investing, but about, because it's America, we are gonna do it this way, right? We're gonna invest in the things that, that Americans want are gonna need, whether it's the railroads here or candy that's sold in the, in the, in the airports here, or utilities, or, I mean, he was looking at, you know, it, it didn't matter if it was tech or not, he wasn't a big tech investor, but he was looking out as what really, what does the, the economy need?

Where is it heading? And how can you position yourself to be at that place when you know, kinda when the puck is, is sent that direction? And you don't have to be, you don't have to be the greedy bastard, I'll use the weird word, who only cares about having the most money compared to all the other greedy bastards.

And that's kind of the, that's the, that's the environment. Whether you're talking about tech titans or you're talking about robber barons, or that was all competition amongst each other, trying to either take the other person out or having the most stuff. Right?

You know, you don't have to do that. I mean, what he, he certainly look, he's on a pedestal above these other folks in terms of this. Here's my point on it.

It's okay. It's okay if that's what you want to be, right? But I, I think we all have to recognize that unfettered market economics, unfettered capitalism without some restraint is not good for the every man is not good for the gen pop, right?

However, neither is some sort of socialistic, communistic thing where, you know, we cut all the corn at the same height, so no one has more than anyone else. Our country has thrived historically as a mix, right? We, we do have market forces that shape our economy and shape our, our daily lives.

But the idea is that through lessons we've learned from the, the Gilded Age, again, they call the Robert Baron age, the Gilded Age, right? Through lessons we've learned through the depression from the Taft Hartley Act, and, and what that wrought through all of these, you know, historical things, we've come to say, Hey, we need a balance. There has to be a balance.

There has to be a, it can't just be, I buy my way in, I settle my a hundred million dollars lawsuit by giving you 25 million, and you wink wink, and let me do whatever I want it. It can't, in order for our system to work, right? We all, or most of us, have to believe that there's some sort of level, level pa playing field that we have a shot at the brass ring, that hard work ingenuity and smarts are rewarded.

It's not, it's not stacked up by some old boys' game, you know, sitting back in the parlor. And all I'm saying is that we had it, it seemed like that's where we were for a while, right? Kimberly, to your point, IBM was on the verge of getting broken up.

Mabell did get broken up, right? Microsoft got hit hard, some might say, uh, for what, what they did with the browsers and windows and everything else. I don't think that's America today.

I, I don't think, you know, and because the other thing I think about it, and, and we will talk about it, and, and another segment, is because we have made tech a strategic national imperative, and we've weaponized it. We're weaponizing it against our perceived enemies. But we can talk about that on another segment of the gang.

We gotta take a break. We're gonna come back and talk about AI anxiety on Wall Street. You are watching Textron Gang.

You've earned it. The spotlight, the responsibility, the weight of teams, companies, and entire industries fall on your shoulders, lives depend on your decisions, your home life included, that work your protected physically and digitally. Nothing gets through your team without a fight.

But in a globally connected world, everyone sees you, including those who mean to cause you and your organization harm. And now home your sanctuary attackers see an opportunity. Your digital front door is wide open.

And what compromises your home can breach your boardroom. Because the devil's greatest trick isn't targeting your workplace firewall. It's convincing you that your personal life isn't at risk.

Black cloak, digital executive protection, defending the new attack surface your personal life. Hey, everyone, we're back. Wow, that was quite a discussion.

Uh, we, it's almost a continuation of this though. But you know, maybe, maybe our friends down on Wall Street are feeling a little of this, you know, we used to fight the good fight too, and they're getting a little shaky and nervous. Mitch, what do you think?

Well, uh, you know, everybody is talking about the AI bubble doesn't know whether, you know, what, what, anything about AI or not, right? It's common. Like, are we on the cusp of having a big correction, right?

Over 2%, uh, two to 3% correction. And everybody's anticipating that. And I think almost that looking for it causes it to happen.

I mean, we're recording this on Friday and had opened the, the markets down four to 500 points. Um, and there's lots of reason for it. But, uh, n uh, Nvidia got hit yesterday, um, dropped in its value.

And, you know, think as if we get, keep on this kind of rollercoaster, things will be pushed back up and then something will draw it down. I think it's, it, it's, we're at a place of saying, is AI real? And how much is, have we over rotated if we've over rotated in investment in data centers and chips and stocks, um, in anticipation of ai?

Or is it real? You know, is it, is it gonna deliver the value and it's gonna, it's, those investments are worth, worth it? And if history does repeat itself, you know, we've over, uh, invested in data centers before in the cloud, early, early in the cloud era, and we've overinvested in other things too.

So it's not unheard of. I think that concern is real, but what's, what's also bolstering it is the job layoffs that don't get reported by the government now, but, and especially in the tech industry. And so we're, I think people are concerned.

It's, it's a time of uncertainty about where this is headed in, uh, as the financial stability of the country or ourselves, frankly, you could argue that's what some of the results of the election mid-year election showed, showed. So, midterm election. So Mitch, you, you know, you said, when that we sort of look for it and that's causing it, and I think that, you know, I looked, I, I did a Kimberly thing, and I actually looked this up, uh, uh, December 5th, 1996, Alan Greenspan gave a speech to the American Enterprise Institute and coined the term irrational exuberance.

And I think what we're doing is not looking for it, but recognizing it now before it becomes really irrational and saying, you know, something doesn't smell right here. So, and I don't remember the exact numbers, but I believe I heard that, uh, open AI has made commitments for a trillion dollars in capital expenditure spend with something like, uh, $30 million in revenue, right? Or 30 billion billion, 20 billion expecting 20 billion hundred billion revenue, right?

So that's just, I mean, that's just such an jack. I think It was only 8 billion, not a trillion. That's amongst friends.

I mean, come on. But it's, it's just so unbalanced. It's like these are unreal numbers that how does anybody make a commitment to spend that much when they have the capacity to pay it back, is just absolutely non-existent.

Yeah. And that's where, that's the irrational supers we're recognizing. I, Jack, I just got a quibble with you saying that it's not irrational.

Maybe, I mean, too technical, irrational meaning divorce from rationality, um, the, what people are spending is based on what the market's at right now, but it's not a rational amount. It's based on what they, it's based on their, their vibe that this is super important. And so in this capitalist, you know, system that we have, that we were talking about on the last segment, um, uh, people pay what they think they should pay, but they, they don't have a clear idea of what the return is going to be.

And I, I don't have the numbers in front of me. There have been recent studies, um, uh, or recent looks at this sort of thing from, from analysts, far more qualified for that than, than I am. Um, say something like, you know, the best return we can expect is something like 20% on the, the, the level of investment in AI right now.

And that's just a fact of the people paying the market rate for stuff without having a clear idea of what, what the return is gonna be. So that's irrational. I think the irrational No, I, I did, I, I'm sorry I did real quickly, Mitch.

I believe it's irrational. I'm just saying we're recognizing it very early in the cycle rather than when everything has already collapsed because of it. Yeah, I think I, what I would take exception to what you're saying is the rational part of it is, is continuing to doing what we're doing, knowing that some of this is doesn't make sense.

com go, go to in an IPO in the ni late nineties for some ridiculous, absurd amount, right? That's irrational, even though it happened, even though people made decisions that could, that, uh, helped that happen. And that's the question, are we in that place now?

Is the overextension of commitments around? Is the buying of buying, buying of business that's happening by co-investing across companies, you know, is that causing, you know, an, an overextension in, in the market and what actually what things are really worth? And the answer's probably yes, and there'll be a correction, and that's what happens.

And, you know, so, You know, the thing I've learned about bubbles over the years, I'm gonna, I'm gonna quote the great sage, Bobby Balala. Tony Soprano's, right hand man. Tony's the Godfather, man.

Okay, go ahead. Tony. Tony once says, Bobby, do you think, do you think you, you feel it when, when they come and hit you?

When, when you get shot? And Bobby says, you probably don't even see it coming. You probably don't even see it coming.

Most bubbles. Last week we were touting here how great this was, and the let the good times roll baby. And then before you know it, that's, that's how quick these bubbles burst.

You don't even feel it. You don't even see it coming. You don't feel it.

And then it's, here is this, that moment I wish I knew I would be out with Warren Buffet somewhere taking it easy, but I'm not, I'm here working. Um, but, but I, I think what we're all saying is we do realize there's a little irrational exuberance here. There might be a payoff down the road, but the way our market, you know, back in the day, wall Street was ruled by institutional money.

It was big, rich people or big companies, institutions that moved stock. Today's Wall Street, 40% of your, of your equity business is retail. It's people like us, not that we're Warren Buffet or anything, but it's small time people like us.

And, and generally, you know, the law of the herd plays in into these things. And that, and that might be what we're seeing here, right? We're, you know what they always say, follow the smart money.

Who the hell, if I knew where the smart money was, I wouldn't be here still too. But you know, that's what they say. Follow the smart money.

Um, Kimberly, I, I know you had, you had written some, some. No, no. I only comment as I'm welcoming a correction right now.

I, I think it's, to my personal opinion is not that I'm a super, I'm not a really good investor or whatever. I, I hand it over to some other people to take care of that stuff, but I believe that we are over, we, we we're over our heel, you know, over our skis right now. And, um, when you look at the valuations, and I think we need a correction, and I'd like to see the heat coming outta the market, and hopefully that's what's happening right now.

So it's like, let's, let's get at some of the heat out and, um, maybe it won't explode if, if we do that. Absolutely. I mean, and the other thing is, you know, hindsight is always 2020, right?

com era. Hey man, I helped take a pub, a company public in that time. We went public in early 2000, or maybe it was late.

Yeah, early 2000, January, February, you know, uh, our Merrill Merrill Merrill Lynch was our lead, uh, you know, broker take banker, taking us out. We had the wonderful man, Henry Blot. Henry was our lead banker.

Of course, Henry got in a little trouble because he was out touting stocks that he was then selling stocks. And we were, we got wrapped up into that, right? I'm not, we're not supposed to talk about it.

I was told never to talk about it, but, um, right. But, um, I think the statute of limitations is run by now. It's 25 years.

com and some of these, you know, there, there was some abuses built into the system. We may find out 10 years from now, five years from now, that there are some irregularities. There are some, you know, things that we needed to correct that would prevent these kinds of things.

But I don't know if we ever prevent them. Totally. I I think it's, it's the herd.

It's the law of the herd, Right? I, I wanted to add something here because I mean, we have been talking about valuation fatigue for quite some time now. You know, that is, again, that's its real story.

And that that is happening in the market. Correction is due to kind of, uh, in due, in due course of time, we will see that. But I think there are two technical repercussions of this.

One is that, uh, the AI data center movement and the underperformance of the revenue, which is kind of coming through, and this is, uh, again, uh, showing up with, uh, what you see as a correction in the market. So there is no substantiate revenue as such, uh, to date for these AI data centers and how it'll play around. That's the first, uh, uh, fact technical factor.

The second technical factor, I would say that, uh, a multimillion dollar revenue wiped out, uh, due to the geopolitical restrictions. So what will happen as a correction is probably, and I think, uh, there is some announcement today that there will be some corrections coming from a regulatory perspective, how this chip market would be normalized or harmonized. Because if the market is reacting strongly to this, this is in making.

So these are two points I would also like to make here. I I think the geopolitical thing, Kareem, is huge. Kimberly, you had sent something else out about, uh, some recent bill passed by the Senate.

Yeah, well, that's having to do with where, who's selling where and, and holding Back. Well, it's, it's geopolitical, right? They're saying, Hey, we are going to prioritize domestic customers over international customers.

You know, you see that during wartime, when I read that you said, Kimberly, I'm saying, yeah, you know, during wars, we've seen that sort of strategic kind of thing. You know, in, in another time, in another place, would an Intel or an A MD or a uh, Nvidia say, Hey, you, you can't tell if they're willing to pay more money, right? I'm not owned lock, stock and barrel by the US government here, right?

And, and stuff like that. So there is, anytime you get a bubble burst and people lose money and get hurt, there's an outcry for regulation. It's just like a knee jerk reaction, right?

And, and I think we're, we're gonna see it here, assuming that this trend does, does, uh, stay the same. Anyway, This point, like 95% of the revenue for Nvidia from China is now zero. Well, you know, we, we've discussed this and we'll discuss it again, but I really gotta take us on a break here, guys, 'cause we're running late and we've got a third segment coming up.

You're watching Textron Gang, Discover Techron Group, the epicenter of tech innovation. We are your go-to for reaching IT, leaders and practitioners worldwide. Our secret impactful content that sparks awareness, engagement, and top quality leads with us.

You'll access editorial websites, streaming videos, virtual events, custom content analyst research, and more. Join our satisfied clients. Let's revolutionize your tech journey.

Contact us today and tell your story to the world in the most powerful way with Techron Group. Hey, everyone, we're back here into our C block. Thanks for staying with us.

Our next, uh, segment is parsing processors. It seems no matter what we're talking about inference and, and all of the things that could come up to displace GPUs, they are the ones present and future kings of the block of the pile. Uh, this is coming outta some futurum research.

Uh, Kimberly, you want to lead us off? Sure. Um, futurum just put out a, um, research report on the market shares of the different, um, chips or GPUs out there, and of course, NVIDIA's at top.

Um, but what is really great about the report that came out is seeing the other guys come out and, you know, why we think, see Nvidia as the big kahuna here. Um, a MD is coming up. You've got a whole bunch of other people.

Um, one of the ones that I've picked up on that's up there is Rock. And I think that's because I just listened to a, you know, a discussion about Humane, who is a, a big, um, investment over there in Saudi Arabia. And their first data center is going to be based on GR and then followed by a MD.

So it's kind of like they started out with the other guys as opposed to the one that we see. But I think the other piece that you had brought up on this, um, is, are these, you know, how, how we're going at it, how this is going, you know, it's gaining ground in the data centers, um, but where is this going in terms of investment and depreciation schedules, et cetera. You know, are we doing the right things in terms of reporting, um, expenses, et cetera, that are out there, um, in terms of the, um, company.

So anyway, I will turn it over to you guys to talk about this. You know, where, and ne next week is, um, super computing. So you, you guys are gonna be right front and center where all this stuff is at.

So let's talk about it. Yeah. This, this week, as, as this airs it's starting this week, sorry, this afternoon, this evening.

Yeah. A, we're On Monday through the magic of quantum time dilation here at Tech Trunk Studios. Go, go ahead, guy.

Um, so it's, it's not just competition with, uh, the kingpin of the GPU nvidia. It's competition of the, the G against the GPU generally with other processing units, so-called xus, npu, neuro processing units, tpu, tensor processing units, pdq, like, whatever it is. Um, this has been happening for a while.

I think the market is just taking time to sort of catch up to this idea. So it starts with saying, which began in graphics, which is where the name comes from. It starts with saying that a central processing unit, a single processing unit is, may not be enough for certain special applications.

Even A GPU is, is, is, uh, you could even call or a, call it a generalized processing unit. It has a, a, a limited instruction set and a certain capability to it, which has been adapted very well to, uh, to, to ai. Now, naturally, these other ones are even more specialized.

What is not often talked about, um, in general, even in really technical discussions, and this is where I'm saying like the market is still catching up to this, is how many advances in chip design there have been over the last 10 years, it to be able to, it, uh, you know, more or less custom code, uh, a, a, a processor to do very specific things and then manufacture those ma manufacture them very quickly, potentially at scale inexpensively. It has to do with innovations in, in what's called packaging, meaning, you know, used to just be one chip. Now that we have these giant chips, but we also have so-called chip lets, where you can collect a bunch of chips.

It still looks like one chip. I don't wanna bore everybody with all the examples of this, but it allows for something like a software development approach to what is a physical thing that gets plugged into a board, which also has, you know, had innovations. But the market remains highly focused.

And the, the, you know, I mean, talking about technical buyers success, highly focused on GPUs and what the future and research report shows is not just is NVIDIA's dominance breaking, but the GPU's dominance may be starting to break as well. And, um, this is, this is really great. That's where efficiencies come from.

That's where people, of course, now the software needs to be developed in parallel to take advantage of these things. That's the next wave. I would also like to add something here.

I mean, these are a couple of things I wanted to highlight. Two things. The first problem with the report, which, uh, Kimberly had mentioned was that the valuation of, uh, these, uh, GPU investments or chipped investments is, uh, skewed up.

Because, you know, if you actually, um, have a life cycle of two to three years practically, but you are showing it in the books to five to six years, that means it's an in inflated valuation. And, uh, there, there are, uh, certain bodies like IFRS and GAP who are regulating this. I think, uh, there is a substantial amount of speculation that how these assets would live that longer life.

And of course, uh, uh, what it means to innovation, right? I mean, there is a lot of like, traction in, you know, how these chips are being developed. And to your point, gay, that what happens is that we are seeing a surge of ps, right?

And PS are like cost, uh, cost efficient and also energy efficient. So, you know, substitution towards G GPUs, but the market has speculations about it. It's not substantiated what value it creates and all that, right?

So, and how it will be adopted. So there are, there is a twofold kind of, you know, movement going on. And we need to kind of watch out how this turns around and how, how it comes across.

Because the bigger problem is that this infl inflated valuation of the assets, which we have in the data center, uh, which will have repercussions that, uh, you know, these innovations, which we are talking about PS and all this kind of, you know, open, open standards and, you know, software like approach to chips would be stifled. So this is, uh, the, the, the crux of, you know, how, uh, this would play around. So I, I, I have a theory.

Let me do this one, Mitch. I have a theory. This is the first generation, kind of the first wave we've seen of, of the, of the AI GPU craze, and I call it the American Cadillac era.

We, we are, we, we use GPUs for everything. Anything you, you want to do something with ai, you need A GPU, right? Because we, we have, we're overbuilding 'cause we really haven't turned the dial enough to understand what we could get away with and where optimum is, we're still dialing in, is the word, right?

I, I think as we go on, we're going to get better and say, Hey, instead of the GPUs, the XP, all these other things are probably more efficient give for the job that I needed to do. Not every job needs the GPU and especially as we move more from training and more to inference, which is what everyone is saying, right? Our reliance on these Cadillacs, people are gonna say, no, I could use a Chevy for this, or a Ford or, or a Toyota or whatever you want to call it.

And I, I do think that's gonna bring some rationality to the chip market Has some good data in the report, Alan, that backs up what you're saying, um, which is the, the primary driver around GPUs is the time to train models. Yeah. The huge investment that goes, that's what drives the biggest purchase of those.

But interestingly enough, the workload is, is starting to change of what people are using GPUs and other X ps variations. So the, the kind of farther down in the report of kind of tuck, untuck, some of the data, uh, it talked about, uh, both training and inferencing was about 33% of the use cases followed by mostly inferencing, which is also at 33% in the same 30 range. Training dominant was really only 19.

And, uh, data prep, the data that we use is kind of a small 10%. But the point being is, you know, it's not just a, a linear curve or a, or a hockey sticker all built around training. Yes, that's happening.

And we, they wanna drive the cost per token around down for training and make it faster, but they also need to drive the cost per token down for infant ring and, and other use cases. And that's happening. So I, it's, I think we're turning that curve that you're talking about.

It's high of infrastructure in making, right? I mean, if you see, uh, the XPU growth is 23%, uh, in 2026, that's the projected, uh, you know, val value of the growth. And then you see how the hybrid infrastructure would also open up new opportunities for smaller players, right?

So that's another thing which we should watch out for. If I could make, And it's worth mentioning also, because, because yeah, uh, Jack, it just worth mentioning, 'cause Q Comms last week that DRA, which is, uh, one of these, uh, uh, Kubernetes related projects just went general availability in August. DRA, uh, dynamic resource allocation name doesn't tell you so much what it does.

It is a prime way for applications to get delivered these diverse compute resources, whether G-P-U-X-P-U or whatever. So the, the software is, is respond The platforms, no, I think we, it's the next wave, but guys, we gotta run all, you know, we try to keep these within the timeframe. Give, give Jack two seconds.

Oh, I, I, if I can just leave our audience with three thoughts, which is, I hate to be the historian here, but we've been through this before. I went through this with co-processor with, uh, you know, uh, co-processor and with, um, the, uh, risk versus SIS scores. Second thought is we do have a change in US finance laws about how you can capitalize equipment.

So we won't be seeing depreciation this year in the us which will change how all of this reports at the end of the year. And the final thought is, despite having this wonderful list of processors that are available now, GPUs, xus, CPUs, et cetera, there are two fabs in the world that make these things. There's Intel and TSMC and that's it.

And that's going to be a very big issue next year And years to come because they don't come on that quick. But speaking of quick, we've gotta quickly end the show. Mitch Guy, Kimberly Garima, Jack, thank you.

Thank you for watching. We've got, got Textron TV following this. Check it out.

But this is Alan Hummel, we're out. Hey everyone, it's Alan Shimo. Welcome to CubeCon North America 2025, or some people call it Cloud Native con, or as I called it, way too long a line to get in here today.

They gotta do a better job during COV when they were checking vaccines, it wasn't this long a line, there was no reason for it. I waited 45 minutes to get in here today, so I don't know if it's an a TL thing or an a TL cloud native thing, but we expect better anyway. Now we started early, so I had some time to kill with that.

Let me introduce you to our first guest here from the show floor at CubeCon. Andy Suman. Andy, you may, if you, if you're a fan of Tech strong learning events and webinars, you might have seen Andy.

He's done more than a few with us. Okay. Um, Andy is with Fairwinds, and don't worry if you don't know Fairwinds, we're going to make sure you know him after this.

Andy, welcome to our coverage. Thanks for being our first guest. Thanks for Having me.

You wait online this morning? I did. I'm glad I made it here in time for the interview.

Yeah, me too. I was, I'm not sure I was gonna make it. It was a little crazy.

Anyway, Andy. Yeah, I said you with, uh, Fairwinds, but tell us a little bit of kind of what your role at Fairwinds is and how you came to Yeah. To that position.

Yeah, definitely. So I, I've been a long time infrastructure guy. I've worked in infrastructures, worked on infrastructure basically since I was a kid.

And, um, about nine years ago at my previous company, I got into Kubernetes, was really excited about it. It was very early days. And then I joined, at the time, reactive ops was the name of the company.

What was the name? Reactive Ops, uh, reacts Reactive Manifesto, which many Cool about. All right.

Little Trivia right there. Yeah. Yeah.

So we, we built and maintained Kubernetes infrastructure for other companies, and I've been doing that ever since, uh, for the last seven and a half years. Uh, working with Fairwinds As part of delivering services is an AWS partner as well. Yes, we are, uh, an AWS advanced tier partner at the moment.

Um, and we've been working closely with them for many years the entire time, but over the last couple years, we've really strengthened that partnership. The majority of our customers are on AWS uh, we are, we also operate across all three clouds, but AWS is definitely the lion's share of that and Very cool. Yeah.

So, you know, I know a little bit about Fair Ones, but let's assume the audience out here doesn't, you guys, I mean, number one, Kubernetes experts first and foremost, right? Some of the Number one thing. Yeah.

Yeah. I mean that's, I, I've listened in on a lot of the webinars. As I mentioned, the, the, the knowledge and skill gap for Kubernetes infrastructure deployments is, is second to none.

Um, but as you said, you work with all the different cloud environments. You work with a law, a big range of software, including a lot of open source tools. Yes.

Yeah. Right. We actually have several of our own that are very popular.

Goldilocks Yep. Is something you guys have have, and it's open source, anyone. Yep.

You don't have to be a, a Fairwinds customer to use Goldilocks. And, and that's yet another interesting thing about fairwinds, right? As you guys are developing the tools that you use for your customer engagements, you actually open source them so that anyone can use them, which is pretty cool.

Yeah. Yeah. We really enjoy, you know, sharing the knowledge that we have back with the community, not just in the form of services, but also in the form of open source.

Uh, Another, I love that One of our major projects is, um, Pluto. So if anybody went, lived through the Kubernetes one 16 upgrade, and all of those APIs got removed, uh, at the time it was very hard to tell if you were still using those APIs. And so we wrote Pluto to help with that.

And I think today it still helps a little bit. It's not as big of a problem as it used to be, but APIs get deprecated and removed all the time. It doesn't change.

So I, I, I wouldn't bet it against it. Not, well, I wouldn't bet against it not being a problem At some point, I'm sure in The future, again, with APIs and AI and MCO server. What about MCO servers?

What are you doing with them? MCP servers? M ccp, excuse Me.

Yeah. Um, not a ton at the moment. We do have an MCP server kind of in the works for our product that we use to do.

Of Course you do. Everyone has One the cost, I mean, yeah, it's kind of table stakes at this point. Yeah.

So our, uh, software product, Fairwinds Insights, which all of our customers use to get insight about their, their, uh, clusters, and they'll be able to access that via MCP here in the, you know, within the next year or so. Of course, MCO came before MCPI guess soon you'll have M-C-Q-R-R. But, um, Uh, Andy, for people who want to get more information on Fairwinds, where do they go?

com, uh, go to our GitHub where all of our open source lives, that's Fairwinds ops is the company on GitHub. Uh, those are the two primary ways to get ahold of us. We all have, so have an open source Slack community, uh, that you can join.

It's on any one of the read mes on our projects. Excellent. Alright.

If you don't mind, I want to kind of pivot a little bit and talk about recent, uh, development, recent announcement with you guys partnering with AWS in the, you know, in the platform engineering space, right? Yeah. You guys are, uh, partnering with AWS on a new IDP offering, internal development platform offering.

Yep. Internal developer platform offering. Tell us about it.

Yeah, so we've always managed the base level of infrastructure, but our customers often need something above that. So it's always been, it's your job to deploy your applications and do your CICD and these days, um, I think there's a need really for, you know, platforms internally for platform teams to be able to deliver clusters and deployments and standard pa happy paths to their developers. And so AWS has built, uh, sort of a blueprint for this with a whole bunch of different open source projects.

So we've got Argo workflows, Argo cd, backstage, cross plane, wow. All put together in this really nice package. And so what we are starting to do is we are offering this as a services offering to our customers.

So we'll come in, set that up, build the, you know, kind of control plane cluster for you, give you all the patterns that you need, help you deploy your first application through the platform, and then hand that off to you. Or we can manage it long term with our standard managed services, but really it's about zero to 60 on a platform in a very short amount of time. And so, you know, there's a few different flavors of backstage.

There's open source backstage, and then we have, you know, uh, I think Spotify still has, They have One, they productized it, it was their, you know, they pioneered it. I'm not, I'm not banging it for it. Oh, no, absolutely.

But, um, how, how does this offering stack up to some of the commercial backstage offerings? Um, I'm not entirely familiar with all of them. You know, there are quite, there's a lot a few out there, right?

There's a lot. Um, but really this is truly based on the open source. They're also working closely with the, uh, canoe effort, uhhuh, um, to really build this fully open source.

And so I think as with all open source, what you'll get is ultimate configurability. You'll be able to do anything with it, you, you want, but it will be a little bit more complex than using an off the shelf solution. Yeah.

So it really, It's not, it's not as pretty maybe. Yeah. Yeah.

And that's where we come in. We're gonna come in and help you make it as pretty. So I think once we are done delivering this open source to you, I think it'll stack up really well against those paid offerings, Really.

Yeah. UI and everything else. Absolutely.

Yep. Very cool. And now that's an offering you're doing with AWS Yes.

As an AWS partner, we're working with them to deliver that. They're the ones who are they picking, like besides backstage? You mentioned there's some GI ops and some other stuff.

They're the ones picking that, or does the customer get to put anything they want in there? So they've picked kind of the core pieces of it, and then there are certain parts of it that the customer will be able to pick and choose and swap in and out because there's an entire, um, off solution built into it. So at the moment, the open source uses key cloak, but obviously not everybody's gonna use key cloak for SSO.

So there'll be plug and play options for SSO for your code repositories. So we'll be able to do GitLab or GitHub or Git or any of the other, you know, kind of flavors of that. Cool.

So it'll be a little bit of both. And then because they have you doing their service, if they want to add other things that maybe even aren't in the core, right? Yeah.

We can add things on, we can help to customize it. You know, everything we do is very, uh, highly tailored to our customers. Right, right.

You know, obviously for operational reasons, we try to keep them similar. We use best practices everywhere. Right.

It's suspicious scalability. Sure. We are a very high touch service, and we really wanna make sure our customers get the, the platform that they need for them that fits them, which is why I think a lot of off the shelf platforms don't work because they're not custom tailored, but then building your own takes so much time.

com days, you probably weren't there. I wasn't working. No, No.

I, I realized this when I was talking to someone. The other, they said, yeah, I was in high school. com days, I helped start a company though.

It was what we call an a SP application service provider and back. It was before this cloud, before this hypervisor, all that stuff. Yeah.

And, uh, we were offering Lotus Notes, which you probably have Heard. I I've heard Of that. Yes.

Lotus Notes, PeopleSoft, Oracle, bunch of like major enterprise Onyx or CRM and stuff. Major enterprise, uh, um, applications. Right.

And the rule of thumb we learned then, it was in 80 20. No one just takes some app like that and just plugs it in and runs. There's always about 20% breakdown session that needs to be.

Our mics are good for that. It's always about 20 cent, 20% of customization that needs to be done. I don't think that's changed.

I don't think so either. In fact, I think the percentage might be higher when you get the open source. Really.

Yeah. Maybe with open source. 'cause it is, you may not have UIs and also you just have a lot more, uh, options.

You know, if you want it to be green on Mondays and blue on Tuesdays with open source, you could do that. Yep. Excellent.

Yeah. Andy, I don't know if we mentioned the website for Fairwinds. Did we?

Uh, I'm not sure We did. com. It's very simple.

Very simple. Yep. Guys, so take it from me here.

Shimmy. com. Andy, it's great to see you in person after, after we seeing you on those little webinar screens.

Same. Thanks. We're gonna take a break.

We're live here at K Con cloud Native Con, we'll be back. We've got a full opening day here, so stay tuned. Hey everyone, we are back here.

This is, uh, I think gonna be our final interview for day one of Q Con and what a day it's been. You know, there's, I don't know, 10,000, 12,000 people here. The, the expo floor is cavernous though, and cold.

Very cold. I'm glad I wore a heavy sports jacket. This poor guy's here in a short sleeve T-shirt.

You're not cold. I'm moving a lot today. So if I were sitting in a chair, you'd be cold.

I'd probably be cold, but I'm moving a lot, so I'm okay. Absolutely. But tonight is the, uh, what do they call it?

The cube? The cube crawl or whatever. Oh yeah.

The coup crawl. Yeah. The, You know, the, uh, They're gonna have drinks, food, everybody's gonna be hanging out in the, In the concert.

Yeah. There's a word for it. Yeah.

But it's cute crawl and, and it's, you know, the, the in expo hall reception. That's right. Yeah.

That's A good word. Yeah. Hey, if you don't know this guy sitting next to me, you probably are not a big fan of Argo or, or, uh, GI ops Octopus, Gi ops Octopus Cube.

He's, he's with us almost every single cube con I've done. And I probably have done, I don't know, 18 of 'em. Nine.

Yeah. This is my ninth year of C***s. Two a year.

Yeah. So you're right there with me. I, The only one I missed was the original one in San Francisco.

I missed that too. And I, ironic, strangely enough, I was like a quarter mile away from it when it was happening, doing else, and No, I didn't, didn't go over there. So I missed that one.

But then Seattle, I think was that next one. Yeah. And then San Diego, right?

Wasn't there two Austin? Remember when it was in Austin and it snowed? Yes, I do remember that.

That's when I realized how big this was gonna be though. Yeah. San Diego is still, well, I, I will say this.

San Diego and Valencia. Oh yeah. My two favorite ones.

San Diego and Valencia were great. Barcelona was good. Barcelona was really Good too.

That might be because Barcelona's a great Barcelona didn't, so I thought, I thought Valencia was a quieter kinder Barcelona. Ah, yeah, yeah, yeah. It was kind of out there.

It was farther away. Yeah. Yeah.

Well, there's a little bit of, like, the hype factory has kind of gone down a little bit because you have a lot of people have moved into the operations phase, right? Yes. With this stuff.

And so in the very beginning, there was such a push to like, we just gotta learn all the things and jump in and figure out what's useful and what's not. And now we're in that stage where a lot of people are like, Hey, we're operating, we're happy. We're, we're scaling.

We're, we're hitting the normal everyday challenges. I, You know, more than me on this, so I'm not gonna pretend to be an expert as you are. But I really feel like, so first of all, at the beginning it was all developers, I felt.

Yeah. It was very heavy developers, guys in t-shirts and backpacks. Yeah.

And Kube was so hard still. It was really hard. Let's face it.

Yeah. It was so hard. And people were not looking for lifelines, but they just wanted to learn more, talk to people like them, do that community thing to figure this out.

I think, I'm not saying Kubernetes is easy, I'm not saying cloud native's easy today, but it's certainly not as hard as it was. Yeah. But I also think, as you said, we've moved from just pure developers to ops people mm-hmm.

To platform engineers. Yeah. To SREs, to security people.

And, and so the audience, I, I would say let's even throw in now data scientists. Mm-hmm. Right?

And, and those kind of folks. So the audience has expanded, the product mission has matured, and the products and and projects themselves have matured. Oh, yeah.

Where I think it's, it's just a more normalized thing. I I still think the passion is there though. Oh, Yeah.

Yeah. No, and there's, there's always new stuff happening. And you know, a lot of the shift in this last couple, last two years has really been about supporting GPU workloads, AI workloads.

So there's a whole isation happening there. But you all start doing a lot of stuff on cost savings, on upper productivity. So That's the efficiency phase, right?

Yeah. You always, first you try to just get stuff to work and then you make it more efficient. And then somewhere right after that you say, oh, we gotta worry about security too.

Um, which is a problem. Well, I, wait, we gotta stop a second. Yeah.

Sorry, I didn't give you a proper introduction. So You don't know. This is Dan Garfield.

Dan Dan has, look you pioneered Argo in a lot of ways in GI ups, right? Yeah. Um, Argo is now of course part of Octopus deploy.

Yeah. Argo is a project, is making Argo The project CNCF. Yeah.

It's maintained by Octopus deploy, uh, red Hat, Intuit, others. And so we have a partnership with a number of different companies that we maintain the project with similar to Kubernetes. Absolutely.

But I do, I do think there is a special fit because Argo the octopus, you got an orange octopus and you've got octopus deploy. We're blue octopus. So we're, we're like, I Got blue Oc blue and orange, uh, go Together, bring me an orange.

Lemme burn you, I guess. 'cause this is the commercial entity. Yeah, that's right.

Yeah. Yeah. So I got the blue one.

But, um, and I should mention, if you don't know, and you, maybe you don't follow this closely at home, you know, um, Argo is the out of over 200 I believe projects now in CNCF Argo was number three right behind Kubernetes itself and, uh, open tell o hotel. Yeah. In terms of like contributions Yes.

And activity and velocity, project velocity, yeah. Number three. And, and o hotel, uh, is so generalized, right?

It, it really touches everything. So it makes sense. And, and Kubernetes, of course is the foundation.

Sure. So those two make sense, but how do people deploy to Kubernetes? 60% of the time they're choosing Argo CD and number two isn't close.

It's down in the 10% range. Really? Yeah.

'cause there's a whole, there's, it's like everybody chose Argo cd and then there's kind of a sea of other things where like, they're using Jenkins and they've just always used Jenkins, so they haven't moved on from it. And they're throwing coops tail applies using that. They may have good reasons to do so, but, but yeah.

Agreed. When we look at the marketplace, that's what we see about 60% of clusters are using Argo CD today. And, uh, and some vendors, some clouds, it's more, but yeah, we, we see pretty big adoption.

So Let me ask another question there. If you are using Argo, are you by definition doing GI ups? You are definitely facing GI Ops Now, whether or not you've actually implemented those principles, you can use Argo CD in non GI ops ways.

Uh, so for example, if you're using, if you're referencing images using floating tags, you're kind of not really full, you're not really embracing GI ops. And you can definitely, um, use Argo CD with manual sync turned on. So you don't have automated reconciliation.

So Argo CD is pushing you towards get ops. It's encouraging you to get, do, get ops, and it's the best way to do get ops in my opinion. But you can also, uh, if you wanna hold a hammer upside down to Hammer in nails, you could do that.

It's just the tool doesn't want you to do that. But you could do that, You know? So I'll give you a life lesson.

As I've gotten older, dad, the tools make the man, and using the right tool for the job is all the difference in the World. That's true. Yeah, that's True.

So you, you know, you wanna keep using the hammer upside down. God bless you. But life's too short to do that.

I I agree fully. Yeah. Yeah.

So I, I wanna get into some of the announcements. You guys have some of the news coming outta this. Yeah.

But before we do, I, I, I want to kinda solidify the, so we spoke about Argo, the partnerships of running, maintaining Argo. Yeah. You are part of Octopus Deploy though, right?

Yep. That's right. And and they're though Octopus Deploy is a maintainer of Argo.

Yeah. And a large part of their business, I imagine is, you know, in, In Argo and Kubernetes. Yeah, that's right.

But Octopus Deploy is in and of itself an entity. Yeah. Talk to us about Octopus Deploy with Argo as part of it, but nevertheless its own entity.

Yeah. So the, the history here right, is you have, the Argo project is created by Intuit. And, uh, they said, this is the way that we wanna deploy software.

There's a number of other tools within Argo that they were using as well. And they said, we really want somebody to take on the mantle of running this thing. And so they looked to us Code Fresh at the time.

Now Octopus Deploy. But, uh, they looked to us to, to be that group. And so we became the first commercial vendor to come onto the project, help it get into the CNCF, help it, uh, go through the process of graduation.

And, um, to take on that stewardship as, look, we're gonna have a commercial interest in this open source project where if you're using Argo and you want to do scale deployments, uh, ar Argo CD really has one job. Look at a source of truth and GI and get that deployed to the cluster. Everything else that's, it is kind of window dressing right now.

What if I wanna manage a change from one application to another? I want promote something. Well, that's where, that's where Octopus Deploy comes in.

What if I need support Octopus deploy? What if I need, um, help doing architectural planning about how I'm gonna roll this out? Octopus deploy is the answer, right?

So, so as a commercial vendor, we have an interest in maintaining that stewardship of an awesome open source project. Uh, and then providing, uh, both tools and services to help you be successful with that project. The great thing is, because we maintain the project with others, Intuit and, uh, and Red Hat Acuity, we cannot any of us take that project and go and just change the license suddenly and say, Hey, we're gonna take a bunch of features out.

Uh, instead we have to focus on something that's gonna work for all of our interests. But that's the, that Keeps us honest, The foundational model. Beautiful.

It's not, you can't decide. I've got a Secrets program and I'm changing the licensing of them. Yeah.

I wanna, I wanna make money on the ui, I'm Tired 'cause I'm gonna provide a, a better version. So I'm gonna make sure the UI sucks, you know, for the user. We, we can't, we, we can't do that stuff.

We don't wanna do that stuff, obviously. No, But, but that, again, look, I've been in the open source game a long time. That is the beauty of this foundational model where that rising tide lifts all boats, right?

But no, no one company can, can manipulate this through their own financial gate. It's a super important point for governance. And it also means that we have a sustainable program for development.

So for example, we offer enterprise Argo support. Yep. Now, our business octopus is really driven by selling our software.

Right? That's where we make our money. And, uh, so the way that our enterprise support program works is, Hey, we're off of you in surprise support, but basically that funds our open source work.

Yep. So, a as we add more, uh, enterprise support, you need to, you need help with Argo cd or something goes wrong in the night with Argo rollouts, you can call us up, we'll help you fix it. Uh, if we need to provide a patch, we'll do that.

But this is the way that we fund, uh, our open source contributions. And of course, that also feeds into one of our biggest customers is the Octopus platform. Because if, if something's not working for a customer, they may not care if it's Octopus or Argo.

They just want it to be fixed. They want one throat to Choke. Yeah.

So we, so that way, um, you know, any of our customers are essentially funding these open source programs and, uh, making sure that they can be successful, which is, uh, you know, it's nice to have Yeah. But it's, but it's really important that you have a sustainable approach to open source. Otherwise you're just borrowing somebody's time until they don't have time to Decide they don't wanna do it anymore.

Yeah, exactly. Or they want to put the squeeze on and monetize. They gotta Change that license.

They gotta, they gotta get, look, I Saw this in security 20 years ago. Yeah. You know, with, and, and look, to be fair, let me just play devil's advocate a second, Dan.

Yeah. Please. Do You get people who put their heart and soul into these open source projects?

Absolutely. You know, single company maintainer, they maintain a community where 96 or 97% or more of the people don't pay 'em a dime. Yeah.

Won't even beta test or report bugs. And it's kind of thankless. Right.

And then you've got, and what really kind of, in my experience, what kind of tips it is, then you get other commercial entities who come by, use this open source tool that you've put your blood, sweat, and tears into. Yeah. And they're monetizing the heck outta me.

Yeah. Taking money away from you in essence. Right.

And I think for a lot of, uh, for a lot of folks, that's where that's the, the straw. That's the bridge too far. Yeah.

Right. And so they, they do do things like changing the licensing or, you know, making it harder for other people who they consider almost like parasites. Yeah.

If you, you know, I don't blame, uh, companies that need to change their license. They're the, they're the only ones maintaining a project and they're like, look, we can't pay to fund this development. We have to, we have to find a way to make it sustainable.

Totally get that. Um, I'm really proud of the fact that we've been able to build this ecosystem with Argo, where we have multiple maintainers Yes. Where we all have good interest.

That, Well, again, that's the foundational model. It's a cooperation model. And it's not just vendors, it's it's End users Yeah.

Organizations into Example. Right? Like, they're not selling Argo services, they want you to file your taxes and manage your money stuff.

Exactly. But for them to be successful deploying their software, they rely foundationally on Argo to do it. It's so it makes sense for them to invest in it.

It's key. Uh, and then it makes sense for us as, um, to be steward good stewards of the project, because ultimately that's the ecosystem that, that, uh, is, is paying the Bills. Absolutely.

Alright, we're running low on time, so let's turn, Okay. What's news? There's a lot of news, uh, in the Argo project.

2 just came out this year. We shipped Argo CD three. Okay.

2. That means all of Argo CD versions two are now out of support. They're gonna start stacking up CVEs and bugs.

They will not be fixed. A lot of people have not yet upgraded to three. This is not Java.

You do not run it for 10 years without upgrading. You move to the latest version, uh, At your own risk. You wanna stay on that old stuff.

It, you know, at your own risk. Yeah. And it's obsolete.

Yeah. And what we're talking about with, uh, the shift from Argo two to three, we made that a, uh, an arc. So we made very deliberate changes.

They would be very easy to upgrade through. And almost all of the behavior that's changed in Argo CD three can be changed back to two X behavior. We have a great upgrade guide to help you do that.

So that's new. That's exciting. We've got new maintainers.

We just added two additional really maintainers from Octopus Deploy focused on Argo cd, uh, who, who just got promoted on Thursday and are now maintainers Congrat to Them. Yep. They've, they put in the effort.

Want him shout out. Yeah. Uh, yeah.

Eugene, he was walking around here. Just a second. You have Guinea, um, as well as, uh, as, uh, Patrick Close, um, who wasn't able to make it this week.

Yep. Yeah. Ghani, what's his last Name?

Uh, den? No, not the, we have a Evgeni who comes on our show once in a while, but yeah, He's, he's fairly new. But, um, we've seen these new maintainers do a couple of really great things.

Patrick Close, he did a big migration in Argo CD to move GI Ops engine back into Argo cd, which is a massive project he worked with Tu it to get that done. Did a great job. 2 because, uh, the version in Get Changed and he was able to track it took three weeks to figure it out, but we got it done.

So won't be hurting users anymore. We're allowed to see that. So that's going on on the, on the community version of Argo cd.

Big stuff happening there. Uh, and of course, as I mentioned, we offer our enterprise support for Argo, our technical account management, where we do proactive stuff, but we also just shift a lot of new features in Octopus to improve your experience with Oh, very Argo. So if you wanna stage out, you know, uh, for example, let's say I've got 10,000 storefronts.

They've each got a Kubernetes cluster, they each have an Argo instance, and I want to orchestrate promoting and managing all those Argo instances. I can do that with Octopus. If I wanna manage deploying new versions of my software and all those versions, I can do that with Octopus and That.

And that's the, so historically that's where a commercial company that's maintaining an open source project makes their bones. Yeah. You wanna scale to that level.

It's hard as hell that you probably could do it at some level using just the pure open source. Yeah. You can write a lot of scripts, a lot of glue, but you've got a great foundation with cd.

She's why you do it. Right. You know, it was the same thing.

Remember back in the Cloud B'S Jenkins days, right? Yeah. CloudBees knew that when you ran, I forgot what it was, four, four instances of Jenkins, you were probably ready for Cloud B'S enterprise.

Yeah. Right. Because that's the scalability that I think are in there.

Yeah. And it's similar. I mean, you could, of course you can manage.

I, I know people that have 50,000 Jenkins instances and they're all 10 versions behind. Yeah. Well, There is that Because they, they didn't take a proactive approach to managing it.

And that's, that's difficult place to be. Uh, but yeah, there's the not only scalability, but usability and, um, user experience, things that we've been able to add. So those are new features that we've brought in to Octopus.

Now for those that remember Codefresh, of course Codefresh is still running ci and we've got these other, uh, components, GI ops things that we're doing. But, um, bringing in a lot of the learnings from that platform into Octopus, uh, as a unified platform is something that we're doing right now. And we just launched that, uh, very cool for Kon and, and people, uh, have been using it and building on it and growing with it.

And it's based on a foundation that's, you know, a decade old at this point, because it has the ability to do all of your traceability and your, uh, governance, tracking, you know, compliance stuff, as well as all of these scalability features. So it is really nice. It's a kind of a peanut butter jelly.

Best of both worlds coming together. I love it. Situation.

Dan, we're almost outta time. Two things I need you to tell him. Number one, people who wanna follow with Octopus Deploy and what's going on there, what's the website?

com. com. Number two, Argo people who are into the, the project, they could go to GitHub, they could go to the CNCF.

Yeah. Or they could go to Argo unpacked. This is our new Argo podcast.

We've done, I think, really episodes now. We have 1500 subscribers. We just lost it.

Oh, it's a big on YouTube. We have 1500 subscribers. That's beautiful.

Uh, which is awesome. And of course you can subscribe to that on your podcast app. Favorite podcast, Argo and K.

We talk through technical issues. We talk through strategic issues, cultural, philosophical as it relates to delivering software, using Argo cdr, using Argo Rollouts. We love it.

Uh, yeah, we're loving that new show. Alan, appreciate you Mrs. Garfield.

He did a hell of a job Here today. I'm gonna awake this time. I'm awake this time.

We're proud of you, Dan. It's good to have you. Thanks.

Good luck. Continued success and keep doing what you're doing, Man. Thank you.

Appreciate it. Dan Garfield here on Tech Drunk tv. We're gonna take, well actually that's gonna wrap up day one.

We'll be back tomorrow with more. I've got some parties to go to. This is Alan Shimmel.

We're out. Hi, uh, my name is Toma fomo. I work for the, as a program manager, uh, of the development tools at the Lyse Foundation.

Uh, it covers a few different topics. Uh, first the Eclipse id, uh, the one most of the people think, uh, of when they hear about the eclipse from Lyse Foundation, most of the time, uh, it's over 20 years old now, uh, but it's still widely use, not so much as an id, but much more as a platform. And that's why we, we still, uh, take care of it, uh, through, for a working group at, at Foundation.

Uh, then I am also in charge of the openly SX working group, which is basically, uh, the, the only open vulnerable marketplace of VS code extension. Uh, this is a good alternative to, to, to the M Microsoft vs code market. And finally, I also in charge of the Cloud F tool working called, uh, which is the place for, for the new generation of open source tooling project.

Uh, it's more than 15 projects. Uh, most of them are using, uh, modern technology web technologies. And I will focus today on two of them, uh, eccl and AI and, and both are really fast growing and very promising projects.

So, so, so let's start. Uh, you know, in my role as a cloud foundation, I, I speak with many developers, of course, all the time, and, and they're working for project in the industry and also from DevOps teams. And when I ask, uh, the question, uh, do you use AI assistant for coding?

Uh, the answer is yes, in something like 50 to 80% of the case. But of often it, it comes with, uh, little hesitation and, and or smile or something that says, okay, but in my company, it's, it hasn't really approached yet, or we're still tr struggling with, with that. And, and that's exactly, uh, the starting point of this talk.

Uh, AI assistance are already part of the everyday, uh, reality for developers. Uh, even if some organization are still catching up. And of course, it, it, it come because it come with new waste and with with new limitation.

Uh, one year ago, uh, we, we, we had this study from, from MacKinsey, uh, based around AI and coding assistance. And the prediction at that time was that by 2 28, uh, 75% of enterprise software, uh, engineer will rely on AI toss, uh, assistance for coding. Uh, honestly, my guess is that, uh, probably if we run the same, uh, study today, those numbers will be likely, will likely be higher.

Uh, but the real question is not adoption. Uh, it, it's, uh, the integration and how we bring these tools into teams and pipelines with the right gateway and things like security and data boundaries and Providence audits. It's, it's where we are going here.

Uh, no, let, let, let me, uh, make a guess. If you, if you are already using, uh, AI coding assistance, uh, it's probably one of the tool on the screen. Maybe I, I forgot some of them.

But, uh, just sharing most of the popular ones on gig co copilot and so on, also known as Windsurf and so on, there are really powerful tools. No discussion about that. Uh, but here is a challenge.

Uh, most of this tool, if you look, these lists are ever closed, proprietary, or if even if they are open source, they are, they are tied to a single vendor. And, and that's the case as the vendor, uh, in that case, the vendor decide how the AI work, where your code is sent, and how much control you have over your, your own data. And it brings, uh, us to two very important, uh, questions.

And, and this is a question that people sometimes prefer to, to, to avoid. Uh, uh, even either you are coding for as a hobby or, or working in a highly regulated and sensitive industry. And here, there are, uh, two simple question.

Uh, first, where those my cut goals? Uh, e even if you're on, yeah, as I, as I said, even if you're not working on top secret government software or very sensitive, uh, you might be building the core algorithm of, of your next startup maybe. Or, um, maybe if it's the business critical, you, you probably don't want that code handing up in, in, in and feeding the AI model and handing on someone else screen.

And the third one question is what I am putting in without knowing exactly, uh, because when you generate code with AI assistance, uh, it could be called linked to a patent, or it could be, uh, something under a receipt restrictive license, like, like JPL or any other. And you may discover, um, probably too late that you, you, you are, uh, when you are ready to ship your product that you, you, you cannot actually use it. And these are not edge cases.

Uh, there are real risk, and probably it'll only grow as AI becomes a normal part of your coding workflow or DevOps workflow. And as code moves faster and faster, um, this is something we, we, we are really taking this. Yeah, this is really to, to something to look at probably.

Yeah, I will be very quick on this one. I think people are probably are already convinced, uh, about why, uh, and what are the drawbacks of, um, of proprietary on signal vendor, uh, tools. First, the security and compliance and many, uh, of these tools require, require something large part, uh, of your code, sometimes you entire code base, uh, to, to, to a model, to something to, to somewhere else.

And in some industries, of course, that's an immediate compliance broker. Uh, for, for DevOps teams, it can, can also raise questions about how code flows through pipelines environment. The second agreement is, uh, uh, I would say custom visibility.

And developers are used to, to tune their tools for their workflow, their language, their specific, uh, domain needs. And with CLO cross, uh, tools, customization is most of the time very limited and, and sometimes simply not possible. Um, third, vendor locking, uh, a failure.

This one is very, very now well known. Uh, the way I see, it's that when you outsource big part of your, of your daily work, uh, to one provider, you also give you, you also give control over your, your productivity. Uh, a change in pricing, a change in licensing, or even an outage can leave you completely stuck.

And fourth, uh, innovation. And, you know, from maybe the most important one here, uh, you know, for four decades, uh, innovation especially, uh, in the development tools, um, has come from open source and proprietary tools take, basically take that away. And, and with it, uh, the ability for communities to, to, to build, to innovate, uh, and to, to make something bigger.

So this is exactly, uh, where, uh, uh, the province that, uh, t and the project, uh, would like to, to present you today is, is designed to avoid. So what is t Uh, basically, I at the bottom, uh, this is, this is a platform. This is a framework.

Uh, ti is a platform. I, I, as you can see on the screen, it's a very, uh, it's very successful, as I said, very promis, fast growing project. Uh, more than 20,000 stars on GitHub and more than 2,500 fourths.

This is an active and large community. Uh, and, and by the, and that's why if you have ever, uh, wanted to join something, uh, meaningful in open source, uh, this is really, this is really a great opportunity. And, and fear ai, uh, brings AI capabilities directly into this platform.

On top of this, you have, uh, as I said, the tier platform allows you to create its, uh, browser base and desktop tooling. And not only Id actually, but the C community, uh, decided to, to develop, uh, an ID called tier id. And it's a, it's a complete ready to use.

And if you are familiar with, uh, with the code, for instance, you, it'll, you will find, uh, really a sta state of the art. Uh, id, uh, people most of the time will install a t for the first time. They even say, oh, it, it looks like it is vs code, but actually it's not.

It's using the semi ter, for instance, mon and or even sometimes have the feeling that it the same tool, but it, it's, no, it's not a VS code. It's not even a fork vs code, meaning that we don't have the constraints of its forks, by the way. It's a really open, free to use, uh, open source and c uh, effort.

And so let me give you a bit more detail about tier and how tier tier ai, uh, ID, uh, uh, uh, breaks, uh, to, to, to, to developer. Uh, and what is really important here that everything is, uh, open and transparent. And, and this is the most important, uh, thing, but it is the most important differentiator of, of this project.

But if you look at this, I, I I'm sharing on the right, you can use any LLM, you know, the LLM, it's basically, it's your, the brain of your AI system. You can decide the model you use and how you use it. You can use it locally, remotely.

You cannot hybrid. So if you want to completely disconnect your system from the internet, you can do it. It's very, it's a very important, uh, uh, set for, for people working in some, uh, confidential or sensitive industries.

Um, you, you can, you can go hybrid of course as well. If you, you, you need to mix your use case beyond that, uh, there's a point editor, which is based on open source library. You can see, edit and share your prompt.

This is a community effort, you know, uh, in if you, if you have ever switched between LLM, if you are family, you are with prompting and you, you know, that you often need to fine tune your request depending on the LM you use. And if you rely on all the vendor or proprietary tools that I, uh, before I mentioned before, uh, you, you, you cannot change anything here. You rely on the vendor to do this.

And here, this is done by you, or you can tune it and you can rely on the community to, to, to adapt to your context and to your, you are using, of course, it comes with very, very, uh, state-of-the-art features, like having a agent. Uh, and you can create your own agent and you can also connect with external system. And I will give you a bit more detail about that.

Now, regarding agent first, uh, yeah, you, you, you have, uh, uh, you have a set of read to use, uh, agents, uh, like coding agents and testing agents or orchestrator to manage a direction between agents. And you can also develop your custom agents. Uh, it's good to know that, uh, you, you can add a new agent, mission-based AI assistant, basically, uh, ever by coding, but can, and you can also add them, uh, using a ui.

So if you are not, if you don't want to, to spend too much time, uh, uh, working complex tasks to, to complex tasks, to, to develop your own agent, you, you also have a ui, uh, uh, UI features to, to do this. Uh, and of course, uh, when you develop new agents, what is very important is to be able to interact, to integrate into your, uh, workflow and your pipelines. And, and this is where, uh, the model context protocols, uh, comes in.

Uh, it gives agents, uh, more context and direct access to po powerful external, uh, capabilities. Uh, so yeah, I think this is very, very, uh, uh, common, uh, feature of AI system today. You have thousands of, uh, NCB server everywhere.

Uh, you can connect to, uh, gate chat systems, like Drive Google Drive or any, you have many corporate and, and external system and services, uh, now available and where, where you can integrate. And then you basically, when you are using a t, uh, id, or even if you are using the T framework to build your old tool, you, you will be able to configure, uh, really easily, uh, the connection to any N CCP server. And it'll, uh, basically tell, uh, let the LLM know that you have, uh, that this new capability, this new external system exists.

To give you an example, you connect it to a GitHub, uh, service, and then you can create an agent that would automatically create, uh, GitHub issues when, uh, the testing agent, for instance, uh, detect fade bug. And, uh, you can imagine whatever use case you want with that. Uh, so with that, the, you are using as now the capability to directly, uh, get the result and, or act and take action, uh, through the, so it's very important to, to mention that feature because this is really the way you integrate your, uh, your features on your, your basically, you know, when you are developing DevOps feature, you, you, you have a pipeline and you need to have this interaction, this integration and tuning capabilities.

And now of course, it's a very complex slide, and I won't give you, I won this one. Don't worry. I won't describe all the boxes here.

Um, what, what I have shown, so used so far is the id, uh, here, this is an architecture drag the description of, uh, the T platform, the T framework. Uh, as you can see, TII framework comes with many other features build, and there are building blocks. And of course, everything here is, uh, again, it's very important that everything is fully open with a license, which explicit, which is explicitly designed to, allows commercial use, that that's intentional.

And, uh, we want, we want people, uh, in this open project to build on top of it and to bring their own tools into this ecosystem. Uh, and basically that, that brings, uh, me to, to, to, to last stage of the day. Um, I already, yeah, you know, if you look at the, at the market and the tools that we have today, I already think we, we are, uh, at a crossroads.

Uh, AI in deliver tools is exploding, but really here we have, we have a choice to make proprietary urban lock or, or versus open or community driven. And if you look, uh, the past two 20 or 30 years, uh, the, the tools, uh, uh, that shaped our industry, uh, succeeded, uh, because they were open. Uh, uh, and so this is exactly, uh, why I, I really, uh, believe that's et I, it's a great project and I'm making this presentation today, uh, coming with freedom, transparency, innovation.

Uh, it's, it's, it's, I would say it's time to act now. It's, it's now because to you, you have so many tools and, and so, so many innovation, so much tools coming right now, it's time to, to go for open, uh, for an open, uh, endpoint source ecosystem. So here is what I would like you to, to, to keep as takeaways issue.

First, as a developer, a really easy, you can try the ai, uh, fee idea. It's, as I say, again, it's free, it's open source, uh, and give feedback. You can adapt to your needs.

And if you want, you can please feel free to contribute, uh, several as a company. Uh, you can evaluate, uh, fee ID as an open solution, uh, where, where you can really have a full control. As I said, I, I just mentioned a few use case, like having local element and so on.

But it's, uh, most of the time very, uh, critical need, uh, and keeping control over your tools, your models, your data is, uh, yeah, I think something, I know some, something which is not optional. And third, uh, as a tool builder, uh, it's not only as an an id, as I said, uh, using Eclipse tier and the ai, uh, uh, is, uh, is a way to create your own, your own tool, uh, your specific tool for your domains, for your user or for your specific workflows. So basically, whatever your role is, um, yeah, there is a way to to, to be part of, to be part of it.

And remember, uh, this is, uh, the time is now. So that was my last slide. I'm ready to hear your question.

Um, you have the links, you have a way to connect with me, so I'm listening to you. Hey guys, thanks for the throw. We're here with Yasmin Rabi, who is COO of Cloud Vault.

And we're having a little chat about, well, whether or not Kubernetes is the defacto default platform for AI or not. Yasmin, welcome to the show. Thanks for having me.

We have seen at least early on that the hyperscalers pretty much use Kubernetes extensively, but in the enterprise, things may be a little bit different. Some folks are using Kubernetes, some folks are using other platforms. What are you guys seeing?

Um, I mean, for us, obviously we're biased 'cause we get involved if an organization is, uh, using Kubernetes. And I think what I see the most is that across every enterprise, there's at least a little bit of Kubernetes. It depends on whether there's a lot or, um, just, you know, in pockets.

Uh, where we see the most Kubernetes is like fintechs. Uh, large banks are mostly deployed on Kubernetes, and then a lot of hosting platforms as well. When you have multiple copies of the same app that needs to scale differently for each customer, uh, we'll see a lot of those types of organizations on Kubernetes as well.

Yeah. In theory, will AI drive more adoption of Kubernetes? Because from what I can see, there's a lot of data processing and that needs to scale up and scale back down.

And it seems like Kubernetes is the natural orchestration engine for that. But are there other ways of skinning that cat? Um, there's definitely other ways of skinning the cat, but what I have seen a lot is, uh, for example, spark jobs on Kubernetes has been exploding.

Uh, 'cause people have a lot of the data jobs they run for like, from anywhere from five seconds to five minutes. Um, and whether or not they should be on Kubernetes, we're seeing a lot of that on Kubernetes. Uh, and I do think it's driving the footprint in the organizations as well.

Mm-hmm. Is there gonna wind up being a Kubernetes skills shortage as a result? Most likely.

I mean, we kind of are at one, uh, even now, uh, a lot of the organizations I talked to, they'll have like one or two experts, and then everyone else is still in that early learning journey. Um, even most recently, the CNCF came out with a, like a Kubernetes AI conformance document, um, where they're starting to lay out what, what are the best practices for being, uh, AI conformant on Kubernetes. Um, and even just to get there, like you need to be Kubernetes conformant, there's some standards and things you need to hit, but, uh, what they launched was essentially a way for people to do their own kind of self-assessment of do I have the must have practices in place?

Do I have the should haves? We'll, obviously, like we'll see that evolve, uh, over time in the next year. But, uh, it's giving people a place to start because I think a lot of folks were like, I don't know what to do.

Like, the question we get a ton is, okay, I'm starting to do AI on Kubernetes, but like, how do I optimize this? Where do I start? And, and most folks don't know where to start.

I think part of the issue with Kubernetes has always been that there's too many knobs to turn. So can we just manage this at a higher level of abstraction where mere mortals can do this and not everybody has to be a certified Kubernetes engineer. That's a lot of what we try and focus on is what are the pieces, what are the knobs that maybe you don't need to tune and you can let a machine tune.

Uh, it's funny 'cause we're, we're an ML company and often that gets grouped into like AI ml and it's like, oh, do you use AI to then optimize ai? Uh, what I like to tell people is like, what we're doing is just advanced math. And if you can use math to solve problems that humans don't need to, for example, configuring Kubernetes.

So how do you set your requests? How do you set your limits? How do you do that for jobs that only stick around for a short amount of time, but when you do them at scale, that starts to take up resources, that type of abstraction for humans, kind of take some of that toil out of it.

And then the, the thing that has, uh, helped us through for decades now is just automation. So rather than having a human go manually configure things, deploy things more with a push button style is just have the automation run it, um, and, uh, run at the intervals that make sense to the organization, but use automation to solve the problem. We've been having this conversation about stateful versus stateless on Kubernetes for as long as I can remember.

Does AI kind of result in a lot more stateful applications running on Kubernetes itself, or is the data being access externally? Um, I think I, I'll say it's, I've seen a mix of both. Uh, we still are seeing a lot of stateful apps, uh, on Kubernetes.

And, uh, whether or not people are addressing the best practices, I think comes to the maturity of the organization. And to your previous point, the skill sets they have in Kubernetes. Um, I think no matter what the technology is, there's always the right way to do it, and not everybody follows the right way to do it.

So you kind of have to build the tooling around anyway that a, that an organization might choose to do it. Um, as you kind of think all this through for a it, where are you seeing Kubernetes? Is it all up in the cloud or are people deploying it on premise?

Because a lot of these AI apps are accessing data that already exists locally and maybe, I don't know, just data gravity just driving this whole decision? That's a great question. Honestly, across our, uh, customers, it's probably a split.

Maybe it's like 60 40 public cloud to on-prem. Um, but we, uh, a lot of folks are doing Kubernetes on-prem, and they'll either do vanilla Kubernetes or OpenShift is, uh, I'd say in the last six months I've heard way more OpenShift than I have probably in the last two years. I, I think the, uh, migration off of VMware is, OpenShift is a nice, uh, first step to, for folks that are moving to Kubernetes before maybe they go, uh, either vanilla Kubernetes or, uh, go into the public cloud.

So, um, we are seeing a lot of on-prem Kubernetes as well. What do you see people doing that you just shake your head a little bit and say, folks, maybe we should be a little bit smarter than that. I mean, I'm, I'm definitely biased here, but like, people setting requests, like manually setting requests.

So our tagline at CubeCon, it's always on our booth and sometimes we're like, should we change this? But it still gets people, it still gets the conversation going. So it literally says stop setting requests.

And obviously, I don't mean like, don't set your request, you should set your request, but stop setting it manually. Um, and that is something I still, because most deployments are in the, like we have customers that have millions of containers that are spinning up and down every day. There's no way a human can do that manually.

So, um, and if you do, you're probably gonna then vanilla spread it across everything and set an average maybe a P 95. Um, so that is the the biggest thing I'll say of like, people should stop setting these things manually. You talked about predictive AI and using that to kind of manage the cluster, but will people also start using more gen AI tools themselves to manage the Kubernetes cluster and hopefully maybe something like an AI agent might explain how something actually works?

Yeah. Uh, it's a great question and something we're actively working on and, uh, I'll say arguing about as well. 'cause um, we, a lot of our customers have been like, yeah, I would love to use a chat bot, for example, to interact with the software.

Not, you know, I don't wanna teach my users how to go in and make changes or use annotations or anything like that. Um, but where, um, some folks are drawing the line is if it's a read action, so like, tell me about my infrastructure. Tell me the last time a recommendation was deployed, how much have I saved?

That sort of thing. Cool. But if it's a right action, they're like, I don't want the chat bot anywhere near that.

I don't want gen AI to then go make, uh, like real changes in my infrastructure. So it'll be interesting to see how it plays out for folks from us as a design principle, we've always said like, minimal amount of permissions, and then you can always add on if you choose as the user to opt in. So we'll still give people access to do things, but we won't have that as a default.

So then the, the folks, the, uh, IT guys that are a little bit more conservative can kind of have control over how, uh, either agents or chat bots then interact with the software, Right? Because I don't think we fully solved this hallucination issue. So even if it's right, 99 times out of a hundred that one time it's wrong will be the most mission critical application in the stack.

Right? Exactly. And the most important thing and, and everything we do is trust.

Once you lose trust, whether it's ai, ml, it's just basic automation. Once you lose that trust, it's really hard to gain back. And especially for the folks that are deploying software in their environment, they don't want their own internal users, the developers to lose trust in them.

So no one wants to get paged in the middle of the night. Um, and we don't wanna cause that to happen. Do you think at some point we might have AI agents that are monitoring the activity of other AI agents and that's how we're gonna have layers of things to kinda, um, minimize any disruptions as much as possible?

I mean, sadly, yes. I do think we're headed towards a kind of crazy landscape. Um, we'll, we'll see where we actually end up, but I like, I think sadly that is a possibility.

Mm-hmm. Um, as we kinda look at this whole thing, is there an opportunity maybe now as Kubernetes goes mainstream to just say, um, can we simplify this in some ways at the, and maybe that's a job for the technical oversight committee, or is there another way to think about this where we just wrap Kubernetes in a bunch of things and there's a bunch of different projects because the TOC community doesn't seem to wanna get involved in that, the management of Kubernetes, if that makes sense? Yeah, it, it does make sense.

I, I think we'll see a mix of vendors that try and solve the problem and the CNCF, uh, at the end of the day, like their resources are limited, so they're trying to do what they can with what they have. And I do think the, uh, AI conformist document is a great first step, um, in this space. And we'll see in 2026 how much of that goes from, uh, like a self-assessment that is optional to more required things that should be in place.

Um, but it's also opening ground for a vendor that has maybe, uh, some, you know, investment, uh, behind it to come in and provide that abstraction layer on top of four challenges that come up. Mm-hmm. It also seems like maybe we're starting to see some separations of concerns here, because early on there was a data Science Tiger team, and if they were lucky, they had somebody who knew something about infrastructure, and more often than not, they were unlucky.

And the data science people ran into the same issues with Kubernetes that developers did. They said, this stuff's hard. I don't know how it works.

And it's very complicated is the inference engines though, are now being separated from the training exercise. And so, um, is that inference engine gonna be something managed more by a traditional IT DevOps team and that's gonna be part of the Kubernetes stack and that's how we'll go forward? Yeah, for sure.

I mean, that's how we do it here. Uh, I'll say formally Storm Forge, um, of course the, the PhD machine learning guys for the ones that came up with the algorithm, um, and kind of set up both the training and inference engines. But now that we're kind of in a run IT mode, and especially for all, all of the data we look at, uh, each workload gets its own unique model.

Um, and so we need various sorts of ML ops, uh, for our inference engine, and it's all run by the, uh, SREs. Uh, sure the machine learning guys get involved sometimes, um, but a lot of that is actually managed, so I'll managed on Kubernetes and it is managed by the infrastructure team internally. And we've seen just, uh, for Storm Forge, we've been around doing like Kubernetes with machine learning for seven, eight years now.

Um, and that skillset, uh, is, there's not a lot of, uh, ML guys that understand Kubernetes and vice versa, but we found it to be critical for being able to do what we need to do, and I think we'll see more of that. Mm-hmm. So you've been at that for a little while now.

What do you know now, uh, after having gone through all that process that you kind of wish you knew when you first started? Yeah, so, uh, I will, you know, obviously credit for the team, they've been doing this for much longer than I have. Uh, I think I've been doing it for three, four years now here at Storm Forge.

Um, but one thing that we all talk about of we wish we knew in the beginning is you can have the most advanced technology, the coolest machine learning algorithms. None of it matters if the software isn't easy to use. Um, because at the end of the day, the people who are hands on keyboard have a lot of things they need to do in a, a day.

Um, and learning how to administrate and manage your software is not top of their list. Um, so I think the biggest learning for us, uh, luckily we, we learned that a couple years ago and, and got to, you know, put that in, uh, to the product. But it's something that is a design principle for us is that it doesn't matter how good the technology is, if it's not simple to use, it doesn't drop in within the ecosystem, technology won't be used.

Hmm. At the end of the day. You guys are really at the forefront of automation in general and have been doing this for a while.

What's next when it comes to IT automation? What can we expect? Yeah.

Uh, so what's really cool, uh, now, you know, being a part of Cloud Bolt is that the portfolio has expanded. So, uh, cloud bolt's, cloud management platform has been around for like 15 years now doing every type of infrastructure. So being able to take our, uh, machine learning and infrastructure automation kind of technology and, and address that across everything is super cool.

Um, and then vice versa, right? Like, uh, the CMP platform today does everything from zero to one. So provisioning, patch management, um, ongoing day two operations and taking Kubernetes and putting that part of it and all the, the pieces that you need to do, like maybe you have to set up a, an OpenShift cluster, for example.

Um, now we can kind of take a step back and look at the entire life cycle, which is really cool at part of Cloud Vault. Um, and then, uh, we have visibility into all the billing data. 'cause we have a finops product as well.

So, one thing that we will be sharing at, um, at CubeCon, and for anyone interested who wants to see it, they, they should totally come by the booth. But, uh, we are collecting the billing data for all the Kubernetes costs and then mapping that into your usage. So you can see how much does this black box actually cost me?

How do I split that by labels, by different teams? How do I wanna share my, uh, all the shared costs, whether it's like idle costs, it's cube system, the things that people struggle with, um, being able to break down that visibility and then, um, go across your Kubernetes estate, uh, is kind of where we're headed and, and what we're working on right now. Um, if you were appointed the, um, Monarch of Kubernetes, what's that one thing that you would decide that we all need to do to change the platform or fix tomorrow that would make everybody's life easier?

You have total control. What are you gonna do? Well, control, I would, uh, get in place pod resizing to ga.

It's the, the thing that everybody asked for. Um, not all applications, even if they should be able to take restarts, can take restarts. And, um, it, it's currently in beta, which is a little bit of a tease because it works in some scenarios.

Doesn't work at all. And, um, I think that'll be a big game changer for people. All right, folks sharing it here.

AI workloads are coming at Kubernetes. There's no doubt. The only question is, is well just how challenging is that gonna be to manage?

And hopefully it's gonna get simpler. Yasmin, thanks for being on the chef. Thank You for having me.

All right. And back to you guys in studio. Hey guys.

Thanks, fed. We're here with Emilio El Salvador, who's vice president of strategy and developer relations for GitLab. And we're having a little chat about how AI agents are gonna evolve in the software engineering workflows that we're about to build and deploy.

Amelia, welcome to show Mike. Thanks very much for having me here. It's, uh, true pleasure.

I think everybody's kind of excited about the idea of AI agents, but we're a little bit perplexed about how this all might play out. And I put this to you this way. So let's say every developer has one, two, maybe 10 of their own AI agents.

And then, um, they're gonna have a bunch that they're gonna have to interact with other developers who have AI agents. And then there's also gonna be AI agents that will be assigned specific tasks on behalf of the team, and they will be monitoring those tasks and then somebody has to manage all of this. So how do you kind of see this all playing out?

Uh, it's, it's a great question. And, um, I think that we are all learning along the way. Uh, at Glab, we believe that the future of software development is going to be in a space where both agents and humans, um, in incorporate with each other.

And what we, we believe that the role of developers, uh, is going to change dramatically. I think that in the past, what we used to see is that there used to be a couple of tools, a couple of frameworks, but now with the rise of ai, I mean that that entire environment is exploding. Every day there's a new tool, there's a new agent, there are new capabilities, and in many cases is that you are even concerned about making a decision because you're just waiting for the next thing that is about to happen.

Every, every week there's this new release. So the role of developers is changing in the sense that the, as you explained before, it's not going to be a, a solo job anymore. Developers are gonna become more like architects, like more like team, um, team leaders, where you are living a team of agents that will do work on your behalf and are gonna allow you to focus on the things that matter the most, which is about solving problems.

I feel that we all know, uh, that developer or coding is just, you know, a third of the job trying to solve that problem and trying to identify how you use coding to solve that problem is a big chunk of that, of that work. Now, developers are gonna focus more time on finding ways to solve that problem and using agents to delegate those tasks. The other thing that everyone who is using AI today realizes is that, is all those agents are not perfect.

I mean, that's like any, any other team member is that you have amazing team members. They are, you have team members that are really good at something but are not really good at, um, many other things. And you need also to understand those technologies, how they work, how they play a role into your software development life cycle, and find ways to use them effectively and find ways to change them.

So the role, uh, will also change in the sense that we see more people acting as the way we call them, you know, the, the guardians of, of your code, the AI guardians, those who, those people who will guarantee that the work that is being done for those agents meet your expectations or your company's expectations from a security, from a quality, from a compliance standpoint. Mm-hmm. And as we see those agents evolving, um, uh, what we believe is, uh, there will be a rise of what we call this idea of meta agents, which is, um, think of it, an agent of agents that will play a more proactive role across your entire software development lifecycle.

It is no longer about you asking an agent to do things for you. It's an agent acting like a true team member. That agent will have an email account and Slack account, probably a phone number, an email address, and you'll end up interacting with those agents the same way that you interact with humans today.

So how will these agents kinda negotiate with each other? And I asked the question because application development, whether it's done by humans or agents, is gonna be a team sport, and different agents will be doing different things. And how will they, um, communicate with each other and kind of determine who's gonna do what, are they just gonna call us to figure it out?

I think that, um, that is, that is a great question and that's why I believe I'm super excited about the role that we can play in, in that, in that new world. Because at the very end of the day, what you are gonna need is that a platform that will enable that collaboration between either human agents or agents with agents, as you said. I mean, development is a team sport.

And now the difference is that in the team, we're gonna have different team players. And a platform like Glab will help. I mean, those interactions to be coordinated in, in a way that deliver business results.

And at the same time, I also think that we might be looking at a world where there are agents that are being deployed to verify and validate the work of other agents. And we're gonna have this kinda layered hierarchy of agents that are kind of working together, but we can't bet everything on one agent or for that matter, one LLM. Um, not really.

I think that as we see the rapid change in this industry, it's not that we are gonna see, I mean, different agents. It's that we're gonna see like different models specialized in different things. And we'll see agents basically overseeing the work of all the agents.

But I don't see a future where humans will not be in the loop. Eventually, there will be a mer request, there will be code that will need to be validated with the help of AI by, by a human. I don't see, uh, enterprise software being deployed with no human interaction whatsoever.

I can't imagine a bank in the US or in Europe deploying a new version of the software without being validated by the existing software development team. They will be able to innovate much faster thanks to platforms like GitLab. But the role, uh, the role of the mechanics of how software gets developed, validated, tested, uh, undeployed will not change.

Yeah, I mean, to your point, I don't think we can hold AI agents legally responsible for anything. And we can't say that the AI agent ate my homework, right? That's correct.

Is that, that, can you imagine that there was a, a big problem in the bank and someone says, well, it was our fault. It was, I mean, agent XI don't see that happening. Mm-hmm.

What's your best advice for folks about how to get down this path? 'cause sometimes, you know, folks will say, uh, luck is the residue of good design. So what should we be doing today to kind of enable this future tomorrow?

Um, well, I think my, my advice number one is that embrace the change. I mean, um, this is, uh, I feel that, uh, unstoppable embrace the change and find the tools that you feel are the right ones for the job and get familiar with them. If you look at the latest, um, surveys and studies that have been published lately, what we are seeing in, in development is unlike a year, a year and a half ago, we are now seeing productivity gains, significant productivity gains.

And yeah, it's true that the technology has evolved, but the reality is that the main change has been that developers are now used to, uh, basically using those tools. They know how to use them. We are moving away from the hype where AI was able to solve everything for everyone.

And now we are coming to realize that AI is an amazing tool, but it's a tool unlike any other tool. It needs to be properly used. If you don't use it properly, it may cause, you know, uh, some damage.

Mm-hmm. So embrace the change, try new tools, and, uh, of course I'm biased. I, I I believe that platforms like GI Lab will help you navigate that change keeping two ideas in mind, extensibility and interoperability, which are the fundamentals of the new duo agent platform.

Mm-hmm. What will be the impact on DevSecOps? And I'm asking this question too, 'cause it, it seems we went from, uh, being afraid of AI tools to now everybody's using those AI tools, but now we have a shadow AI tool problem, and everybody's not quite clear what they're using those tools for.

And the output can vary greatly. Yeah. And I'm, I'm, I'm glad that you're asking that question because precisely everything that we have been working on is based on that, on that, on that question.

The idea of as, as these new technologies, new agents, new models, we need to find a way to bring all those together in a way that don't compromise the way you, I mean, develop software. And that's what we're bringing with GitLab and the dual agent platform. So you can use the tools that you want, you can use the models that you need in order to, uh, achieve, uh, the work that you need to deliver on, but keep that in a controlled environment.

You need a platform that will bring you full visibility on what's happening across your entire software development, life, life cycle, and will maximize your opportunity to deliver innovation faster. Mm-hmm. Do you think early on we may be obsessed with just using the AI tools to write code, when it seems to me most of the bottlenecks that exist in our DevOps workflows have very little to do with the writing of the code.

And there's all these tasks that we need to do today, and there's a lot of manual effort. It takes the joy outta software development. And frankly, for a lot of folks, the fun part is actually writing the code Could not agree more with you, but not all.

I mean, it's like, but then when we look at the stats is that developers are spending less than a third of the time writing the code, which is the thing that they love the most. And it's precisely because of that, because they are, I mean, either writing, volley, play code, writing documentation, finding bug, and I feel that ai, well, I find AI is helping with that. It is optimizing the time that developers are spending on doing the things that they like the most.

It's not just the coding, but it's the problem. Solving, coding for the sake of coding is not fun, but it's like when you apply your skills to understand the problem and solve that problem with code, that's when magic happens. Mm-hmm.

What do we need to do to our core CICD pipelines to enable all this? 'cause there's one wagon recently told me, he said, basically, yeah, we're running more code than ever. So we're basically running into the same wall faster.

Yes. But is, uh, the same way, we're also using ai, not just for writing code faster. We're using AI to optimize your pipelines.

We're using AI to validate the security or compliance of your software. So embracing AI is not just, uh, as part of your coding, uh, coding, coding part of the job, it's embracing AI across your entire server development life cycle. Uh, even if you're able to produce much, I mean code much faster.

And, um, the problem to your point before is that there's always a bottleneck. Is that how fast you can deploy that software? How fast can you, um, secure your software?

How fast can you, um, fix a bug when that happens in production? And that's why it's critical not just to embrace or use AI in one, you know, in one of the, in one of the stages of your software development life cycle, you have to be strategic and mindful about using it across your entire software development life cycle. 'cause otherwise, your productivity gains on one side will become, you know, livity losses on the, on the front end.

Will we need to revisit the pipelines themselves? Because, let's be honest, um, a lot of them today are fairly brittle and they kind of break often and turns out we have more, probably more of them than we actually need. So do we need maybe longer term fewer, better pipelines or just more pipelines?

I think that's the, the, the, the beauty of using ai like the same with code. Do we need more lines of code or fewer lines of code, or we need, uh, one framework or another? I think every application, every environment is different.

I think AI will help you, uh, analyze what you have and optimize that challenge that, that we have with pipelines in many cases, is that they got so huge that, uh, some people, even the people who wrote them, don't even understand them anymore. They, with the help of ai, you will be able to truly understand what's going on, if there's a problem, fix it without you having to go through your entire pipeline definition to understand what happened and eventually optimize them. Mm-hmm.

What's your take on vibe coding? Where does that fit in, in the DevOps workflow? 'cause in one sense, it's a tool for citizen developers, and they might be creating code, but not sure about the quality of that code.

And does that mean that all those vibe coding tools are now essentially pushing code into some sort of C-I-S-C-D pipeline that another set of software engineers and their agent buddies are gonna have to review? Uh, um, well, there's, there's like, like everything else, right? It's, there's positives and negatives with that.

I feel that by coding is also helping accelerate the idea generation. We've seen companies that used to write every single spec, and now it is much easier to explain what you want to develop just by, by coding it so people understand truly what you, you, what you meant instead of writing an spec. And the other thing is that the problem is people may believe that those tools will create the code by themselves, and then it just, you know, things are gonna just work.

And we know that development is more than, than just that, especially when it comes to enterprise over development. And yeah, I think they are good in terms of helping developers initiate the process, but those tools are also becoming more and more complex and more and more professional. And you can also inject, um, custom rules, um, architectural guidance.

So the vibe coding experience is going to be, is going to get better and better and better. Yeah. Some folks will say that we're about to build and deploy more software in the next two years than we might have in the last decade.

Um, that's an optimistic view of things. But, um, I, will we get to that point? Or is it gonna take a little longer because we got optimize the pipelines?

But, you know, it, it's like most things, it's a journey and we always underestimate the amount of time it takes to get to one place, to another. Um, it's a journey, right? And, uh, as I say, it's a marathon, not a sprint.

I see what it's happening today is that companies are able to reduce technical depth significantly. I think they're able to innovate much faster. I don't know whether that's gonna translate into, uh, much, um, much more software than, than than we had before.

I think what we're gonna see is software being, uh, adapted much faster than we saw in the past. Mm-hmm. And to that point, are we gonna focus more of our time and energy on new applications to replace the probably many of our existing legacy applications?

Or will we go in and kind of modernize these legacy applications? 'cause the time and effort to do that might not be as intense as it once was. I think it's gonna be a little bit of both.

I happen to be with a global system indicator the other day, and one of the challenges that they face when they speak with customers is that they're dealing with massive applications, hundreds of lines of code that they can't touch because the people who develop those applications are no longer in the picture. They retired a while back ago, and now what they're facing is that, well, we need to modernize those applications in order to innovate instead of replacing all that code, which is almost impossible. So I feel that we are gonna see a bit of both.

We're gonna start seeing more applications being modernized, and once they get to that point, people will keep working on those applications, make them better and better. Yeah. And then the one thing everybody seems to be trying to figure out is there's no shortage of new vendors showing up with AI coding tools and platforms and these things.

And then there's the traditional existing DevOps guard, um, who's gonna win this fight ultimately, because, you know, on one hand the upstarts have the quote unquote cool factor, but on the other hand, companies like yours have all the data, Right? And I think that is the critical part that the idea of having a single data store that will provide companies full visibility of your entire software development life cycle. I see a world where those two things will coexist.

There will be newer tools that will help developers do things better and faster. But eventually, as I said before, server development is complex. It's not easy.

It's not an app that you develop, you know, to run on your desktop for yourself. When you're thinking about enterprise grade software applications. You're gonna need a platform that will give you the, the security and the consistency, compliance and that, that companies need.

And that's why I see a world where Glab will become the core platform that will enable those scenarios where developers will use different tools. AI providers will bring different models. Uh, agent developers will come with their agents, but you will need a true orchestrator that will bring all those capabilities together.

All right, folks, you heard it here. The one thing that is certain is DevOps and software engineering is gonna change exactly how and precisely when remains to be the termed. Amelia, thanks for being on the show.

Mike, thank you very much for having me here again. It's been a pleasure. Thank you.

All right. And back to you guys in the studio. Vast weaves a deal.

The vast crypto spectrum, HPE decides to go its own way. Red Hat makes some updates to OpenShift. AWS is under the sea.

Pentagon is farming out ai. And we're gonna take a closer look at some of the announcements from KubeCon 2025 in this episode of the Tech Field Day Rundown. Hello everyone, and welcome to the Tech Field Day rundown.

It is November the 12th, and we hope that you're having a very happy lunch hour or whatever hour it happens to be, because it is national Happy Hour day. Who knew that, that that was a thing? We're actually having a very happy hour here at Commvault Shift in New York City.

Uh, it's, it's a lovely blustery day, uh, just basically like it always is. And joining me, of course, is my co-host for this episode. Stephen.

Stephen, it's good to see you again. It is Nice to be here. I haven't been on the rundown in a while.

Uh, those of you who, um, follow such things, you can mark that down on my scorecard, uh, because Tom and Al have been handling it so capably, but it's really nice to be back here for the rundown. It is because we've got a lot of news and some of these things are definitely right up Steven's Alley. 17 billion.

It is a major step forward in AI infrastructure. The multi-year deal makes Core Weave Vasts largest customer, and expands the use of Vasts AI operating system, which unifies storage, caching, and data services to support large scale GPU clusters. As AI workloads grow, this agreement shows that fast reliable storage is becoming just as important as any of the compute power experts now expect storage to make up between three and 5% of total AI infrastructure spending, which is signaling its importance in the Gen AI era.

Steven, you are intimately familiar with Vast Data, and as AI continues to grow, how does this deal help Vast Data position itself in the industry? Well, uh, obviously, uh, by having vast, uh, sign such a major customer, such a critical one for ai, it gives them a real leg up in terms not just of, uh, being part of the AI infrastructure space, but in sort of cementing their future in that space as well, because essentially they instantly become one of the most important, uh, components of the stack. I expect that this will be a, uh, this will continue to be a, an accelerator for Vast Data's growth.

And, um, frankly, uh, just on the face of it, it's nice to have such a big customer. Uh, the flip side of, I suppose, is that it makes vast even more reliant on the AI infrastructure world and on Core Weave specifically. Uh, but I guess there's worse, uh, partners to connect yourself to.

And, uh, frankly, if the, uh, checks are cash in the money's flowing, uh, sounds like a good move for them. One of the things that caught my eye of this announcement though, is that it's not just vast supplying, uh, storage to support Core Weave, it's that they're gonna be using some of the special features of the Vast platform for this. Now, way back when Vast announced that they were gonna be more than storage, and they said that they were gonna have data, they were going to have, um, a, uh, event broker that they could use to, uh, support, uh, applications with structured data.

Uh, they had Kafka, native Kafka inter interface. My question, uh, for Vast at the time was essentially, that's cool technology, but is somebody really gonna use this? And is it really that much better than just using Native?

And the answer is absolutely yes. In fact, uh, vast claims that they're about 10 times better at handling Kafka requests on the same hardware as just running Kafka na natively, which is really a major advantage. And again, the importance of this shouldn't be understated.

It shows that vasts pivot away from traditional storage and into data applications. And AI integration is not just functional, but it's really beneficial to customers. And so, for that reason, I think this, um, says more actually in terms of the benefit to Vast than you might think.

Um, you know, even beyond the fact that they now have a billion dollar customer, Tom Spectrum Labs Spectrum Fusion platform let's organizations verify their cyber resilience using cryptographic proofs. This system tracks key tasks like backup and recovery, while AI agents monitor workflows and provide real time updates, the approach helps it teams demonstrate readiness, reassure boards and insurers, and improve recovery from cyber attacks. What's your take on Spectrum Fusion?

If you ask me what the most important thing about the blockchain, what its legacy is gonna be, it has absolutely nothing to do with any magical pretend money. And it has everything to do with the immutability of the ledger, because one of the things that we need more than anything else in it is the ability to say for a fact, the thing did or didn't happen. And what this platform is allowing you to do is create a token that you can then share with auditors and insurers that say, for a given point in time, we met the requirements that you put in place for this.

Now, why would that be important? And especially when you start talking about things like cyber insurance, well, I don't know if you're the underwriter for one of these policies, you're gonna wanna make sure that the company that you are underwriting did everything they could to prevent whatever it was from happening, from happening. And this token kind of serves that purpose, if you will.

So if I am an insurer, I would say, I would want you to use this platform. What you do is you go ahead and install it. You run it, and it gives you the token.

And the token says that, you know, on, I don't know, November the 12th, 2025, you met all of the requirements to be insured with our policy. And then two weeks from now, someone comes out with this massive zero day that, that nobody could have foreseen and violates all of your security policies. Well, as an insurer, I know that at that point in time, you met what I needed for this.

That's an easy, uh, claim to, to file, right? In fact, that's one of the things that their AI agents will allow you to do is file that claim automatically. Uh, if you've ever worked in insurance or you've ever had to deal with insurance, you know how big of a deal that can be.

But it also allows you to define what things you are going to be, uh, running these token proofs against. It's not just, do we have a firewall? It's not just do we have separation of policy?

It is, are the backups working? Have they been tested? All of these other things that I would want my system to be able to prove beyond a shadow of a doubt that you're capable of doing, because that way everybody knows, and because they can be shared, because they can be rerun at any time, they help provide that proof.

And this just leads to more resilient platforms. That's really what we want. Insurance is there when the resilience fails, but if we can prove that it was resilient at some point in the past, then we know that gaps that we need to address and cover so that everything works out well.

So I, I'm kind of happy to see this. I'm also kind of happy to see a, uh, solution that involves using cryptography that isn't about Monopoly money. Steven, we had news this week that HP is gonna stop selling Qumulo scalability and WCA software by March of 2026 to focus on its own storage pro platforms like Electra store.

Once Zerto and GreenLake, the change simplifies HP's portfolio and strengthens its in-house offerings. Ity and WCA will continue working with HPE through channel partnerships, though the move might impact their sales. And we know that HPE has been working really hard lately to kind of simplify their offerings because they do offer quite a few things to their customers.

What does this move signal for the industry? Well, um, maybe a little less than you might think. Um, fundamentally, HPE has always been a good partner for many companies in the industry, including, of course, companies like Scale a Scale, culo, and wca.

Um, in fact, I believe that Ity and HPE, uh, really sort of blazed a trail there. I think that HPE must have been S scale's first Major, OEM, their major channel and, and frankly helped build that company. Similarly, you know, WCA has had just tremendous success in the AI space.

Uh, you know, Qumulo has a long, long history of developing great products as well. And of course, HPE has an OEM version of Vast Data as well that they sell in addition to their own, uh, electro storage platform. Uh, this change essentially is, uh, HPE saying, you know what?

We should really be focusing on our internal, um, products and, and a little less on some of these partner products. But that doesn't mean that it's a fundamental change. In fact, I suspect that Channel Partners, I mean, HPE obviously is one of the biggest channel companies out there.

I think that their channel partners are gonna continue selling wca Ity, Qumulo and more on HPE hardware going forward. So I, I really wouldn't read too much into it. Instead, what I think this is, is it is just sort of a cleanup of the way that HPE goes to market, the way that HT HPE works with, uh, both partners and OEMs and the channel, and shows that HPE is increasingly confident in their own internal storage platforms.

So overall, um, I don't think this is gonna have a huge impact on customers. I also don't think that it's really gonna have a huge impact on the companies that were named here that are, uh, allegedly losing access to HPE customers because it's not like those customers are gonna switch to something else. They're probably going to be working with HPE and, um, you know, WCA or Scale or Qumulo for the long run.

Um, I, I think the one interesting aspect here is that HPE apparently is going to continue working, uh, with their OEM version of Vast. Now, I don't want this episode to sound too much like a, a cheerleading session for Vast, but boy, that does, uh, lead some credence to the fact that, uh, vast is sort of the golden child of the storage industry right now when HPE is continuing to work with them, even as they are separating themselves somewhat from some of these more traditional partners. 20 adds new AI tools, stronger security and better Edge support.

Key features include AI workload support, post Quantum Cryptography, zero Trust, identity, and Easier Secrets Management. This update also expands ARM and Oracle Cloud Support BGP networking and two Node Edge deployments, helping organizations modernize applications and manage growing Kubernetes environments more efficiently. Uh, what's your reaction to Red Hat's OpenShift announcements and the overall strategy that Red Hat has for OpenShift?

I think it's important that this came out when it did, because it signals that Red Hat is not giving up on OpenShift. I know I was looking at my travel logs from, uh, years past, and it was only nine years ago that Steven and I were at, uh, the, uh, OpenShift Summit, uh, OpenStack summit down in Austin. And of course, OpenShift is, is definitely related to all of that.

But I think that this is, if Red Hat's gonna offer cloud in a box to people, it has to be a modern cloud in a box, right? You have to be willing to take on all the things that are available in, uh, Azure, in AWS, in GCE in any of those platforms. And the fact that they're supporting Oracle Cloud will tell you that, oh, it looks like Oracle Cloud's actually coming along with all these deals that they've been making.

So looking at the features like post quantum cryptography is kind of table stakes right now. We've seen this shift with a lot of people where they realize that the time to get in front of this problem is right now, because the implementation of, of post quantum algorithms is not gonna cost you anything right now. It will cost you a lot whenever those data, uh, storage units are encrypted with older technology, and you have to burn CPU cycles to get them upgraded.

Now's the time to do that. Of course, you know, you want to talk about things like zero trust secrets management. Yeah.

That is, again, table stakes, right? You, you have to have some kind of an isolation mechanism for these workloads. You have to be able to do secrets management as we learn from all of our friends that have done a number of presentations at Security Field Day.

Uh, like one password, for example. If you can't do Secrets management, nobody's gonna rely on you. Again, adding arm support, adding BGP support, adding two node edge deployments, these are the things that I would expect from a modern platform.

And I know that the people at OpenShift have been working really, really hard to kind of be the alternative to doing things in the public cloud. They're your private on-premises cloud. And, and this gives me hope that this is not something that's just gonna disappear.

Obviously with Red Hat and IBM behind it, it won't, but this can be positioned as an alternative for people who are under massive regulatory strain, that that cannot be in the public cloud or can't have the majority of their data. And that's especially true for people who are sensitive about having their, uh, data uploaded to AI clusters everywhere. So, good news from Red Hat, if you're an OpenShift customer, I'm sure you're gonna wanna upgrade pretty soon because there's a lot of new bells and whistles that you're gonna wanna try out.

The Pentagon is pushing AI adoption in the military, but they're expecting companies to pay for the development and train that AI themselves according to Defense Secretary Pete Hegseth, tech and defense firms, including Google, AWS Meta, Palantir and OpenAI are going to have to provide these contracts to the Department of Defense. But the basic training needs need to be done by the companies themselves. The focus, of course, is on fast acquisition using commercial technologies and deploying AI in things like drones, autonomous vessels and other systems.

The companies are supporting the effort, but experts are warning that they're gonna have to balance speed with quality and of course, shareholder expectations. Steven, do you think that the Defense Department is on the right track here, making the companies do all that training themselves? Or should the Defense Department pony up if they expect the models to be trained?

Well, I think that, uh, this is a little scary to think about. Um, you know, you, you don't necessarily want to chat GPT deciding whether to shoot you or not. Um, or at least I wouldn't want that.

Uh, but on the flip side, I think that this is more about trying to figure out how we can leverage, uh, commercial technologies, as you said, for military applications and less about, uh, giving Claude a gun. Um, you know, that being said, uh, this is fairly typical for, um, government to work with the private sector in various different ways. Uh, typically when the government has a, uh, job that needs to be done that's specialized that the private sector can't have anticipated, they do, uh, support that in many ways, including paying for development, uh, or building in, uh, you know, building it on a cost plus basis so that the private sector can meet the needs of the military that are usually dramatically different than civilians.

Uh, I think that ultimately what we're gonna see here is, um, applications of the, of AI technology that are very similar to what you would find in the civilian space. Um, this is not going to necessarily be, um, you know, using a chatbot to do some things that you would maybe not, it would not be prepared for. But that being said, there are a lot of companies out there that are actively working on developing military applications.

Uh, you know, we've of course talked about, you know, uh, the big guys, you know, anthropic and AWS and Google Cloud Meta, Microsoft, Oracle. Um, Palantir is one company that has put in a lot of their own development resources to support the needs of government and the military. And of course, there are others, um, specialized companies that are developing drone technologies.

I was, uh, talking to somebody who works with one of them, um, in Ukraine, uh, recently. And, uh, they are doing just remarkable things with AI and drones, um, terrifying but remarkable things with the drones in Ukraine, uh, using AI technology. And I imagine that that's pretty attractive to the US government as well.

So really, I think that what this means is, uh, the US wants to be able to leverage AI as quickly as possible with in as many places as possible. And hopefully they will approach it in a way that is, uh, at least technically going to result in, uh, a, a usable and not, um, Skynet type product. Tom, uh, Amazon Web Services has announced fastnet.

Its first fully owned transatlantic sub-C cable. It links, uh, Maryland to Cork County Ireland. It's set to go live in 2028, and it will deliver 320 terabytes per second of bandwidth, which is enough to stream 12 million HD movies, which apparently is our metric for bandwidth.

Now. Uh, this strengthens AWS's global network and supports growing cloud and AI workloads, uh, the 4,000 mile cable ads redundancy as well, uh, outside of traditional routes, improving resilience against outages or damage. It's built with advanced optical switching technology, and it reflects, uh, AWS's push to get more control over their own in infrastructure in-house.

Uh, what's your reaction to fastnet? Sounds like a fastnet to me. It does, and this is a big deal for Amazon because this is their cable.

They are not renting it from anybody else. They own this thing from Maryland to the island of Fastnet where there is a lighthouse and assuming some other things that are kind of awesome. Uh, I don't know if they're gonna put the data center in the lighthouse, though.

It's Kind of a cool coincidence that the island is called Fastnet, Isn't it? Like, it's almost like they did that on purpose. Uh, here's the thing that I worry about with Amazon, though.

Uh, one of the reasons why you don't see a whole lot of privately owned under sea cables is because they're hideously expensive. Uh, I did not notice in the article did Jeff Bezos get a Navy, because he's gonna need one in order to lay this cable. It's 4,000 miles long.

That is several meters from my back of the envelope math. But what's really important is the fact that it's gonna be armored, right? Like you have to armor this thing to avoid, um, rogue NAS from dragging their anchors on the sea floor and taking it out, or, or who knows what else.

But that is not a cheap process, and you're gonna have to lay all that cable. I'm sure if they started now, they'll probably do be done in like three years when it opens up, Amazon's gonna have all of this massive bandwidth to still make people use US East one as the default instance. Uh, I, I think what they need to do though, is they need to start selling this as you can start hosting these, uh, you know, production units and everything in Ireland or in European data centers.

And, you know, you can move data back and forth as quickly as possible. Yeah, 12 million HD movies sounds like a lot, but how many AI workloads can you move? Because one of the things that's cited in the article is the fact that AI workloads can get awfully bursty whenever they find a new source of data and bandwidth to, to jump into.

And they're gonna have to do some really impressive things as far as like kind of metering all of that. I'm curious to see how, if this is an investment that pays off, of course, if I had Jeff Bezos's money, uh, I have to do something other than launching celebrity singers into space on rockets. Um, this is probably something a little more pragmatic that Andy Chassis is really gonna appreciate, because it's gonna increase AWS's bottom line overall.

All right, we, uh, had a couple stories we wanted to take a closer look at, because this week is kon, it's in Atlanta, so we hope that you're all enjoying some peaches and, uh, Braves baseball. Uh, but there's been a lot of announcements that have come out of, uh, KubeCon that we're definitely gonna wanna check out. The first is that the Cloud native computing foundation, CNCF has introduced the certified Kubernetes AI conformance program to standardize how AI and ML workloads are gonna be running on Kubernetes clusters.

The program ensures workloads are portable across different environments, including hybrid and sovereign clouds, backed by companies like everybody's favorites, Google, Microsoft, Oracle, Broadcom, and Red Hat. It aims to prevent platform lock-in and simplify large scale deployments with AI workloads increasingly managed by centralized IT teams. The initiative supports the magic word, interoperability, scalability, and efficiency through deployment across Kubernetes ecosystems.

Steven, do you think that the CNCF has enough clout to standardize how we deploy AI on Kubernetes? Well, I think they have enough clout to standardize how we deploy on AI on Kubernetes. I am not sure they have enough clout to standardize how we deploy ai.

And I think that that's an interesting and important distinction to make. Essentially, uh, AI applications often run on, uh, dedicated platforms that, uh, have no Kubernetes in sight. Some of them do, some of them don't.

Uh, when they do have Kubernetes, uh, many of them have that somewhat hidden from the implementation, uh, from the, the, the higher level details. Some of them, uh, allow the customers to see the fact that it's, uh, that, that all that's running. Um, I think that ultimately what this is, is it's a reaction by the CNCF to the fact that one of the most important, um, new workloads is not necessarily Kubernetes native.

And the companies that you mentioned, uh, all have bet big on Kubernetes as the future of, uh, infrastructure. Uh, and so, uh, they wanna make sure that they're ready for that when it comes to AI applications. So, uh, if you were cynical and if CNCF was a corporate entity, you might say, well, this is just a, a ploy for them to maintain relevance.

But CNCF is not a corporate entity. In fact, CNCF is a pretty remarkable industry collaboration. And I feel that in fact, this is not a cynical approach.

I feel that this is, uh, an example of a need within the industry for a, uh, portable way to run AI workloads. It's a need or a desire within the industry to have AI and cloud share similar infrastructure. And ultimately, I think it's good for the industry.

And it's the sort of thing that CNCF ought to be doing. Hi, it's me, cenic, but not for the reason that you're thinking. I'm actually cynical in the opposite direction because I love that CNCF has decided to do this.

They're like, we are essentially the stewards of Kubernetes. We need to keep Kubernetes relevant because, well, it's basically the cloud right now. What if you're running something in the cloud, it's probably running on Kubernetes.

The problem is that they see where the future is headed and the future is crashing through the streets like a bull in a China shop, because AI doesn't care. It doesn't care about frameworks, it doesn't care about platforms. All it cares about is consume more, more power, more ram more storage.

And that doesn't play well with things that have frameworks around them because there's a way that we do stuff. And I think that C NCF is basically saying, if you're gonna use Kubernetes to do ai, you've gotta make it play well with everything else so that it doesn't just basically act like a swarm of locusts and consume everything in sight. I wonder though, if the companies that they are targeting with this, everybody's favorites, Amazon, Microsoft, Google, Oracle, are gonna play well with that, knowing that if they make CNCF mad, they might lose future input into the way that Kubernetes works.

Because I got news for you folks. AI isn't the only workload out there. There are a lot of things that are still gonna be running on containers long after whatever the new hot thing in AI is, is put out to pasture.

So if I were the companies and, and I know you're watching this, uh, Larry Ellison, uh, char Andy Jassy, all of my friends don't make CNCF mad play. Nice. You can still have access to all the ram you can get your hands on.

Just, uh, just go with it a little bit. Here's another announcement from KU CO 2025, and it's from our friends over at ARM because they highlighted their Neo verse platform and Google Cloud's new Axion in four A VMs, which are enabling energy efficient, scalable AI and cloud workloads on wait for IT. Kubernetes collaborations with CNCF partners like Harbor, OPA, Edify and AuthZ help developers build secure, portable cost effective cloud native systems from edge to cloud.

The focus is on smarter, more sustainable architectures for modern applications. So we just talked about CNCF putting all this framework in place for Kubernetes, and now we have ARM saying, guess what folks? We're ready to adopt it.

Think this is a good move for arm. Oh, yeah. Um, yeah.

ARM has obviously, uh, come to dominate mobile computing. Uh, they are taking a serious stab at dominating, uh, desktop and client computing as well. Uh, they're an awful lot of ARM devices in the Edge network, and frankly, they are heading to the data center and AI as well with this, with this platform.

Um, ARM is very, very smart to recognize that the thing that keeps people from selecting different hardware platforms is often not hardware, but software. I mean, the hardware has to be good enough, but the, uh, software also has to meet the needs of, uh, developers and of end users. And that's exactly what ARM is doing here.

Essentially, they are working with the industry in a productive way to make sure that everybody can use their platform and that leads people naturally to want to use that platform. Now, another aspect of this, of course, is the announcement that by Google that they're gonna be having their new Axion N four a, uh, cloud VMs, um, built on the Neo versus platform. Uh, we have seen, um, the proliferation of ARM-based instances in the cloud, but generally those haven't had the impact that we might have thought they would.

I think one of the reasons for that is because customers honestly don't really care about performance per watt. They just care about performance per dollar. And to this point, that has been a bit of a stumbling point.

It's sort of blunted the impact of arm cores in the cloud. But if ARM can do more than, uh, the competing X 86 based platforms in areas like AI inferencing or, um, you know, other areas, if they can deliver the goods at less money, I think that that could be something that could be exciting to customers, especially if it just works. And that's the challenge.

I think it is too. And I love that ARM is embracing this because, like you said, the optics around, uh, reduced cost per watt, per whatever don't really matter. The electric bill is what matters.

And the electric bill is a component of how much I'm gonna rent these things out to it. So if ARM really wants to make an impact in this, they have gotta figure out a way to cut this cost to the bone to make these things way more appealing to companies that are saying, well, where should I put these workloads? 'cause if it's all running on Kubernetes, honestly, nobody cares, right?

We're gonna compile it on whatever platform we need to use it for this week, but if you can consistently deliver the things that make it a lot cheaper, then you are gonna want to embrace those platforms, right? We saw this originally with X 86. It won because it was more efficient and cheaper than all the other options that were out there.

And then it became the lumbering behemoth that it became. And now ARM is kind of sneaking in to take some of that steam. And if you don't believe me, look around your desk and realize that with maybe one exception, everything you are running is running on an arm core, whether it's your mobile device, your tablet, your laptop, and even if you have one of those X 86 laptops that's running a Redmond operating system, there's probably an arm core in it that's doing a lot of AI offload.

So Arm is kind of the way that we're looking at this. And I quote, everyone's favorite, 1995 Seminole movie, hackers Risk Architecture really did change everything, and it appears to be poised to take over the world. Uh, congratulations an Angelina Jolie and Johnny Lee Miller, you you live in our hearts forever.

One more thing I wanna mention here at the end, Tom is, uh, we are at CubeCon with, uh, tech Field Day this week. Uh, that's one reason that, uh, we are, uh, recording these things at a different time of day. But the other reason is because you and I are not a cube con.

We are here at Commvault Shift in New York City this week. Uh, Commvault's announcements unfortunately are embargoed past the date that we are going to be releasing this episode. So I want to give a shout out to Commvault thank you for hosting us, and we, I promise we will cover your stuff next week.

Um, next week is also the day, you know, on, on Wednesday that, uh, Commvault's going to be live streaming their, um, or not live streaming, but streaming their, uh, announcements from shift. And we're gonna be doing a live blog of those announcements on the Tech Field Day website. So, uh, we are very excited to be here at Commvault Shift this week, but, uh, we are not covering it on the rundown this week, but we will cover it next week.

We take embargo seriously. Mm-hmm. We, on the other hand, have had a number of other Tech Field Day presentations from, uh, CubeCon this week.

Again, uh, the timing of the recording and production of this means that we weren't able to cover those on the episode this week, but I'm sure that we will be talking about some of those things next week as well. So, just a brief program note, if, if you're wondering why it is that, uh, we're not talking about Commvault Shift and, uh, why we didn't cover the Tech Field Day at CubeCon, that's why. Yeah.

And make sure you tune in for that live blog on the 19th, because I'll be sharing my thoughts about all this cool stuff that I know about that I can't tell you about yet. But trust me, you're gonna wanna tune in. com for a lineup of all the events that we have coming up, because guess what, folks, we're getting close to the end of the year.

You know what that means? We're already planning for 2026. We have a lot of things in motion.

We have a lot of events that are already scheduled. com, you can see the list of those. I've got some stuff going on, so you has some stuff going on.

Our good friend from down under Alistair Cook has some stuff going on. Find the event that works the best for you and put it on your calendar so you don't miss out on that opportunity. Just like we know that you have a reminder to tune in to watch the Tech Field Day rundown every Wednesday, we love making these episodes for you.

We love finding all the cool news that we wanna share with you, and then we post it to YouTube, uh, or, you know, maybe your favorite podcast application of choice. Uh, there's lot of input that's been going into Apple Podcast, overcast recently, and we love to hear all about it. But we want you to tell everybody else what we do here.

That means leaving us a rating and a review, all the stars, all of the thumbs up, tell people how cool we are, because if you do that, then we get put in front of more people who love to hear about the ins and outs of enterprise tech news. Um, that's, uh, an underserved market. And, and I think we're doing most of the serving there.

Uh, don't forget though, that Steven and I also have a lot of other things going on with, uh, other future and group properties. I have a podcast that's related to Security, security Boulevard. We just released a new episode with my friend, ed Whedon.

You are a regular, uh, guest on Textron Gang, along with my good friend Alan, who's a co-host on Security Boulevard. And we just launched an AI podcast with fu with, uh, Futurum, uh, and Textron as well. So look for utilizing AI in your favorite podcast application.

It's me, uh, Nick, patience, Olivier Blanchard, um, Mike Fazar, et cetera, talking about, uh, AI in a similar vein to your Security Boulevard podcast. Absolutely. We're gonna be back next Wednesday with all the news that was in the it week that we just finished up.

And we'll be talking about Commvault Shift as well. Uh, make sure that you tune in. But for myself, for Steven Foskett, for Matt Garvin, who's been hanging out in the back, making sure all the cameras are straight and all the audio levels are recording properly.

Thank you very much and we'll see you next week. He's Really here. Stop pooping.

He's really here. We're not, we're in the same room.