Exploring Cloud Native Adoption and Platform Engineering with Brent Schroeder – SUSECON 2025

This is Textron tv. Hey, everyone. We're back here with our continuing coverage of S Secon 25 in, uh, sunny Orlando, Florida.

We've had some great weather this week. Um, our next guest is, uh, Brett Schroeder. Yep.

Brett is chief of the office of CTO. I Head the office of CTO, head, head, the office of CTO at SUSE Worldwide. And, uh, joins us here today after, I guess you've had a busy three days of speaking to customers.

It's very, Very busy, which, which is exciting. It's, I think this is the most exciting, uh, sussan that really we've had since I've joined seven years ago, so, yeah. Good For you.

Yeah. You know, I'm trying to think. I've been to a previous Sussan over in Europe.

Was it? I don't know if it was Berlin. Berlin Might have been Berlin.

Yeah. Um, you know what, what I loved about this event this year was, look, it wasn't one of these mega events with thousands of people where you feel like you're just, you know, on the subway or something. Um, it was a place where you really had a lot a chance to really have great discussions like that hallway session.

Yeah. You know, the water cooler sessions were really va for me anyway, I had a chance to meet and talk to a lot of people. Brett, a big part of your job is meeting and talking to people.

Yep. Right? Yep.

Why don't, well, I don't want to describe your job. Okay. I don't You describe your job.

Yeah. So as officer as head of the office of CTO, probably 75% of my time is spent talking to customers in, in one manner or another. Um, so that's why I'm on the road.

I pick up mail in Austin and, and, uh, the rest of the time it's on American Airlines. Yeah. But, um, so it is, it is about that.

Right. And, and, and how I start off almost every conversation is tell me about you. Mm-hmm.

Right. What are you experiencing? What are you trying to do?

What's holding you back? Um, because I need to learn, right. Um, what are their problems so that then we can apply technology and, and, you know, hopefully bring them the biggest value, uh, the customers, the biggest value from our portfolio that we can possibly deliver.

So that's my role is, is really gonna, I Love it. Um, you know, one of the things that I've been, I, I actually wrote an article on Cloud native now about it, and it, I've been talking to the people. I was very impressed with this DevX design Yeah.

That they, they showed us over in the solution showcase the other evening. I was, I got a hands on tour with it. Um, it's very exciting because I think as, as we were talking kind of off camera, I think one of the issues around cloud native adoption, especially at scale Yeah.

Right? It's that could do science experiments all you want, but at scale is it, it it's almost like drowning in a sea of opportunities. Mm-hmm.

There's so many different solutions for every single facet of a cloud native deployment. Yes. That I think people are almost, you know, paralysis by analysis that Yeah.

No one wants to make the wrong choice. Everyone wants to pick what they perceive as What's their tool set of, right. Favorite tool set, tool set of choice, and why it's better than the next one.

Yeah. Yeah. But you know, this, this Dev x, uh, uh, framework to me was so comprehensive uhhuh, and, you know, and it really showcased all of the different pieces that Zu has assembled here.

Yeah. And to pull it all together into a platform. Yeah.

Truly a platform like that. I can't wait to, you know, I gotta imagine the market's gonna just love it. Yeah.

I'm wondering what your hearing see. Yeah. The, um, you know, this really comes back from our two to our discussions with customers, right?

Mm-hmm. As I go out and talk to customers, there is no one model. There's no one IDE, there's no one language.

Uh, and so as we sit back and think about that, it's, you know, us coming out with another, you know, hard opinionated, this is the way you must do things, would not resonate with the vast majority of of customers. And so, and, and going back to your point on scale, you know, I think many customers are now wanting to pass that point of the science experiment, the departmental, um, you know, initial trial, right. And scale.

And so customers are in running into not technical issues so much, but organizational issues and productivity issues in scaling. Mm-hmm. Uh, you know, how do they grow from 100 developers to 2000 developers?

And if you ask every developer, you know, Hey, do you want to go program some, uh, some infrastructure? It's not too many of them that are infrastructure. Well, because they're, they're not busy enough.

They Yeah. They're, you know, waiting for somebody to send 'em some more requirements, right? Yeah.

Yeah. They wanna focus on the business. Absolutely.

And they wanna focus on code. You know, one of the things I, I remember a couple of conferences last year, the average developer spends about 25%, 27% of their time actually coded. Exactly.

Yeah. So it's telling them they've gotta be the security person, they've gotta be the platform person. Yeah.

Op guy, the tester. Yeah. That's not what they wanted.

No, it's not what they're paid to do either. Right. A lot of them, that's all fine and dandy, but their manager says, how many lines of code have you Yeah, Exactly.

Committed. Yeah. Yeah.

Yeah. I see a very stark, uh, contrast between companies that have adopted some form of platform engineering Yes. Versus those that there's an operations team, but the, the infrastructure as code falls on the developer.

Yeah. Um, they run into the challenges of how do I educate them? I get inconsistent implementations.

Yep. Which lead to inconsistent security, uh, practices among the developers. And so the platform engineering really helps, uh, unify the approach as it transitions from what the developer needs to do and is trying to do from a business standpoint.

And how do you put that into operations, uh, and therefore helps scalability from the organizational standpoint. I get it. You know, so just a quick plug, we Yeah.

com. It's our newest site. It's been up there now for about four or five months.

org community, and Luca and Ente and those folks. We, we were, we are a media sponsor will be a platform con, but that you, you hit the nail on the head there. Right?

com too. So we're obviously we're bullish DevOps, but we saw the same things happening in DevOps. I used to call it the bubbles of DevOps.

Uhhuh. You would have a little team here, a little team there, like bubbles. Yeah.

Within a bigger organization. And this one was on Ansible. That one used chef, this one used Puppet, this one used Jenkins.

That one used harness, that one used Git or GitLab or what have you. And then someone at the enterprise level would say, Hey, we gotta consolidate this. Yeah.

The mess. Right. And it was kinda like, you know, when your kids play with bubbles, when they're little and you put two bubbles together, you got a bigger bubble.

And until eventually the bubble bursts, but DevOps had a problem putting all those bubbles together. Yeah. Because They don't fit.

Right. Well, just for what you said, everyone wants to use their favorite tool. Everyone has, you know, their way of doing it can't just keep shifting left and throwing it on the developer.

Right. I think platform engineering is a great Yeah. Answer to that.

Exactly. It's not that it replaces DevOps, you still have the whole DevOps thing. Still do the, the, but you need a platform that's gonna bring together the developer, the tester, the DevOps guy, the sre mm-hmm.

All of these people that work in this, you know, software factory mm-hmm. That we enterprises run to. Right.

Um, so I I, I agree with you a hundred percent. What's the feedback then? Um, I think it's going to, it's, you know, we're just releasing Devex now, right.

I think it's gonna be fantastic for the customers that I've talked to, and particularly the ones, uh, the ones that are like a couple years down the road on platform engineering. Yes. They're gonna love it because it's gonna give them, you know, the prescriptive, how do I just do the next thing faster?

Um, how do I get more consistency between, um, you know, differentiated teams, uh, et cetera. Those that haven't, it's now gonna give them part of the recipe for how to get started on platform engineering. And that's really, you know, what we're doing with DevX benefits both communities, the platform engineering teams and the development teams.

I agree. Uh, and platform engineering then can say, okay, here's some best practices that I want you to follow. Here's the tool sets that are pre-integrated and, and, and blueprinted on, on how to follow them and how to use them.

So it it, you know, trying to kinda bridge that gap of I need my tool set that fits the problems I'm trying to solve. You know, I might wanna use a different language if I'm doing video, um, applications and gaming Yes. Versus if I'm doing ERP programming.

Right. It's, those are different tool sets. Sure.

And so how do we give them the choice to do that? Um, you know, by giving them best practices and pre integrations with some of those tool sets, but not dictating which one they use. Choice happens.

Yes. Choice happens. Right.

And that, and I think that is the key. Yes. There are a lot of choices and no, we're not here to dictate mm-hmm.

This is a complete set. I think one of the nice things I like about the whole DevX thing is here's the complete A to Z, you wanna substitute a different lettuce seed. Mm-hmm.

You know, substitute a black, you want to use that over there. Go ahead. Yeah.

You do have choice. You're not locked in. Because that's another thing we hear Yeah.

From enterprises a lot, is they wanna stay away from lock in as best they can. Yeah. Right.

Because things are changing and happening so fast. Yeah. It helps you evolve faster, right.

Is is if something new comes along, Gotta have your Options that has a productivity boost, you wanna be able to take advantage of that and not be in a five year transition. Exactly. Plan Or, you know, like what we see going on in the, uh, in the hypervisor VMware world now where we're at sort of a generational juncture where people are actually saying, Hey, maybe there's a good time to modernize.

Maybe it is a good time to go to, uh, not multi threaded microservice architecture. You know, I'm, I'm looking at maybe going from OnPrem to cloud, private, public, whatever. I'm changing hypervisors.

It's a, it's, no one wants to be locked in at that moment in time. Right. You gotta have your options.

Yeah. Speaking of innovation though, and changes ai Yeah. Whether we're talking generative ai, agentic ai, I'm sure you, your, your customers that you're speaking to are grappling with, is it real?

How much is it real? When is it gonna be real? How should I make it real?

Yes. What are you hearing? Yes.

Um, I'm hearing that, and I'm gonna break it down into two, two aspects. Is it Gen AI and agentic ai, I think are, man, this next wave that's gonna open it up to so many more people. Uh, the interesting thing that people have forgot about is that, you know, machine learning has been around for a while.

Yeah. And we already have, um, hundreds of customers that have employed, um, AI in, in just astounding ways, in medical, in, in retail, in manufacturing, et cetera. Well, I think this now opens up is probably the, um, a new wave of how to apply it.

Um, but I think, and, and everybody's looking for it, right? And, and the the headlines you hear every day are where's the ROI? Right.

And I think the key thing to remember in, in the gen AI space is it solves a specific set of problems just like the last ai or that still their ml, um, solves a specific set of problems. Don't try to apply it to every problem that your business has. So my guidance is, you know, really study the business processes, um, the workflows that you have, um, and does it fit and optimize it?

Right. I use Gen AI every day. Mm-hmm.

Um, and it is absolutely a productivity boost to my day. Uh, and I love using it, but there's other tasks that it may not benefit. Right.

So, you know, knowing where to apply it is key. And then as we move to Gen AI or Agen AI Gente, Now, I think that opens up another whole Can worm opportunity. Opportunity, but it can worms too, And Yeah.

Issues that you need to grapple with from a governance and, and yeah. There, and this is one of the things that, that, um, it becomes so important in age agentic ai, uh, your gray areas of how the quality of your data, um, is no longer an option. If you're doing age agentic ai, your data must be clean.

Uh, hallucinations are not a good idea in an agent AI workflow. Did You have that? Right?

Probably not safe to say. Safe To say, Yeah. Should we launch this direction or that direction?

Right. You think to know, uh, Uh, so I think that's the exciting part of age agentic AI is, is what it opens up. But again, okay.

Do I have the problem understood and do I have good clean data to, I think that's gonna be, that, that is a key piece of it. But I, I think the other thing, you know, we've discussed this on text on gang in a lot of our videos and interviews, is we, we can't go hog wild and have a hundred agents running, you know, doing, you know, how many agents are enough agents? Do we need an agent orchestrator at some point?

Yeah. Do you have a master agent? Do you, you know, what's the right mix?

Yeah. Because everybody's rushing out. I mean, SAP's coming out with a lot of AI agents, sales framework sports, obviously.

Yeah. Everyone is. Yeah.

And so, And kill mes coordinate with each other. And how do you get it? They, they're gonna have to, otherwise they're gonna start colliding like, you know, bits on the internet.

Right. Um, there's going to, you know, just the way you have this Dev x uhhuh, uh, platform or framework, we're gonna need sort of a framework to deploying and managing agents Yeah. At scale cloud native infrastructure or whatever your infrastructure is.

Yeah. You know, that because these agents are going to be constantly, I think fine tuning, fine Tuning and tweaking. And, and, you know, you bring up a good point, and I, and I think this is where agents have an opportunity to be disruptive.

'cause we've in it, you know, if we apply automation to it, right? Yep. It's been the holy grail forever.

Right. Automation is always a good thing. You never hear them say, oh, no, automation's not good.

Um, And the challenge with automation is, uh, you know, what's the quality of your information? And I give an example, um, you know, and not picking on it, but just an example, like, uh, VMware, DRS right. Automatic, um, load balancing and re uh, positioning workloads, et cetera.

Yeah. And it operated on an incomplete information, uh, information set. Yeah.

And so in the early days, it was as dangerous as it was value on your book. Right. Uh, I remember those days.

And that's the thing that we're gonna have to watch as we progress with Agent Rolly as good as your information. Right. So it's very easy, like, you know, you can think you're optimizing something and yet incomplete information.

Um, and it was the wrong way to drive the workflow. Yeah. So that's, we're gonna need some guardrails around, we Look at the world through a, you know, a door hole, a pinhole.

Yeah. And you don't realize what else is out there. You wouldn't know.

Right. Exactly. Excellent.

So, well, Brett, thank you for stopping by. Absolutely. Talking with us today.

Yes. Enjoyed continued success with Susa. We're gonna watch all this.

We're Excited. It's the most exciting time. That's why I said it is, it's the best Sushan I bet to, because the, the bringing the portfolio together in a way that we've never done before.

Absolutely. You know, they're on the horizon. Or, You know, as, as a media press person, I've, I've seen the acquisitions.

I I was good friends with Chang from Ah, Yeah. From Rancher. From Rancher.

Mm-hmm. I, uh, Andreas, that's another good friend of mine. I knew the new Vector people pretty well too.

And you know, they're assembling a nice, stable, I don't know how this all fits together, but now you see how it all fits together. It's coming together. So.

Excellent, man. Excellent. Appreciate, appreciate it.

Appreciate it. Yes. Greg Schroeder, head of the office of CTO at SUSE here at Sussan.

We've got more Susa uh, coverage coming up. Stay tuned.