Haseeb Budhani, Rafay Systems | KubeCon + CloudNativeCon Europe 2023
Gartner predicts that 80% of software engineering organizations will establish platform teams by 2026 and 75% of those will include developer self-service portals. In turn, platform engineers are using Spotify’s open platform, Backstage, to build internal developer platforms to help teams move quickly while staying within company standards. However, this is a lengthy and costly process—it’s challenging for platform teams to give and govern access to dedicated and shared Kubernetes clusters and namespaces. Rafay CEO Haseeb Budhani will discuss the rise of platform teams and Backstage. He’ll also discuss why developer self-service is critical for Kubernetes management and environment provisioning.
Transcript
This is texturing TV. Hey everyone. We're back here live.
In I almost said Barcelona. We're back here live in Amsterdam. For kubecon called nativecon.
There was one in Barcelona way back when I'm pretty sure before Valencia before covid. But in any event, I'm happy to be joined by hasibutani. Got it, right?
Yes, sir. Fantastic. We see him every other.
Yukon it seems right we didn't make Detroit but I think you were in Valencia and invite me. Otherwise, I'd been there. You were waiting for an invitation.
It was an information. Okay. Well, I'll just tell you we don't do invitation every time absolutely again, it's good to see you.
I see not everyone. Here's gonna remember rafay or what they do. So I'm gonna ask if we could start there give them a little Baseline.
What what is Rafael? Yeah, so we're in the kubernetes management space. And I said kubernetes management not kubernetes and the distinction is that enterprises are choosing kubernetes as the you know, orchestration platform of the future.
They're for the containers applications and then they'll choose Amazon kubernetes or Azure kubernetes or you know, Google's kubernetes or openshift, right? These are platforms that exist in the market the challenges that one Enterprise's decide to go with kubernetes, then they build a platform organization that writes automation for you know, provisioning of infrastructure upgrades application deployment policy management access management backups multi-tenancy Network policies. Oh my God, that's a lot of stuff that didn't happen so that you can consume kubernetes in the right way and your developers can move fast this platform layer.
We sell as a product so that my customers don't have to reinvent the wheel again and again, and again, they can buy it and be up and running in production in the month versus take the next two years and build up platform. That's a value that Rafi brings to Enterprises. Love it.
com or rafae dot CEO. co just like it says it on there and it's not CEO. Okay.
We got that out of the way now, let's focus in on This is 10,000 people. We're back. but you know that we're not no one's mentioning The Big C word anymore.
We're not wearing masks. It's all good. Nevertheless macroeconomic conditions have kind of changed the market.
Absolutely and you know, there's and any time this changed this winners and losers. It's just the way of the world. You guys have been.
kind of capitalizing on these conditions explain to us. Yeah, absolutely. So a big value that we deliver was valuable before the economy turn and it is valuable now, which is look any of the day Alan the way I think about our business is we are in the people business, right?
The family solved were roughy don't ever exist. It would be solved by people. The challenges that there's not enough talent in our industry.
No, right. So if every Enterprise in the world needs 20, whatever number of people 10 20 30 people depending on the size of the organization man, there's not enough talent in the market, right? I mean, we it's just it's not possible.
So in any case there was need for automation so that we could augment this teams. So these Enterprises could go faster. So that's what our business was when the economy was awesome.
Now the economy is not not as doing not as well and many of these Enterprises are looking for a way to solve the same problems. But they're looking at their TCO right? They're looking at their budgets and they're saying wow.
This is like, you know building a massive team is really expensive. I have the talent I need here. How can I leverage my existing talent and do more?
So what we're doing is we are essentially again same thing for the existing teams. We're telling them with rawface software. You can now move a lot faster with kubernetes and you don't need to add on more headcounter your team because the reality is even if you want to leadership is not gonna let you add head count in this in this economy, right?
But still the projects need to go on right and truly. This is how I think about it like beer augmenting teams. In fact Allen.
I say this a lot I say when you work with us, it's as if you got hired a guy named profit, right? You're gonna hire people anyway, you're gonna hire me. I'm just gonna bring my 200 people with me and we're gonna solve your problem for you, right because and this is the most fundamental kind of fact now, Every Enterprise Healthcare Financial Services high tech they all have exactly the same problem when it comes to kubernetes automation.
Why are we all building it when we could buy it? I I don't disagree with you at all. Now sometimes though.
There's no doubt. People are the most expensive. Piece in the equation and oftentimes the most difficult to source.
Nevertheless there is some organizations that they just they don't Embrace automation as that for good right? They don't embrace it there. They're Skeptics.
How do you overcome that skepticism? Yeah, you know, I think the the great part about this industry is that our customers are actually and they're really smart people. You know, they're all they're all really smart people, right?
And they're they make data driven decisions, right? So They want to try it out. So what one of the things we've done here is you know, retail our customers or prospective customers.
Hey, look, I don't know whether our product can solid wrong for you or not. I have no idea only you can determine whether this is going to make your life better. I cannot I can only tell you what it does.
Right and then you're gonna decide whether it's today. Today's the right time or six months or never. It's okay, but the only way you will learn is you know what I'm gonna it's a fast product.
I'm gonna give you access take as long as you want. All good. Kick the tires.
Here's some documentation. We're Not Gonna bother you and you know by building a model where we provide this kind of no pressure engagement model that I find has worked really well for us as a company because again this these aren't very technical very smart people. They don't like to be pushed.
They don't like to be pressured into buying for example, or even a POC so long as you are very honest about what is possible and what is not I find that, you know, people are making some really really good decisions and the reality is not everybody's going to be my customer I get it. But most will be yeah, right, and that's great. absolutely, and it is a great thing and that makes you know, I always say.
In uncertain Economic Times, the next great companies are absolutely forged. Let me go in another Direction a little bit. You can't walk more than three feet without tripping over AI today, right ji this generated of that gpg this GPT that What what role is that playing in the buffet Automation and a universe?
Yeah, look today doesn't I because we've been focused on building what we're building and you know, we have to be on a passer on but there's two angles we see in in terms of how something like chatgivity could help us and our customers. I'll tell you both of them. So one is we have Sort of audiences inside an Enterprise that are not technologists like finops is an audience for us.
Right because we have it a chargeback engine in a product that tells the customer, you know this team cost you this much money Etc. But those people have sometimes very interesting questions and I don't have an a priority dashboard for them because I've never thought about the question. They're gonna ask what if that data could be made available to them and they could Converse and collect the data they want that's a great use case because we have like we have an API interface.
We have a user interface. We have a CLI interface. Why not a can you know, like a conversation interface that are so much sense.
Absolutely. We should do that. And we will by the way.
Hey just make sure you give me credit. Yeah, because it was here. Yeah.
Yeah the Allen concept. Yeah, and the Second Use case is we have you know, young Engineers who we hire so we hire a you know, a lot of really smart engineers in India, you know, you're young you and their careers it takes them time to learn our platform. So before they're productive it takes six months before they productive.
Yeah, what if the the mundane stuff that they have to learn we could sort of it just comes along. And they don't have to learn it. So as they're writing code some of these things just come along.
Every kind of cool right which means if I could cut I don't know a quarter of time of learning. So that's very yeah, right. It's a lot of money right that to me is a very interesting use case.
So these are the two use things we think about in the company, you know, and what I tell the product management team is, you know, don't get caught up on the hype. That's because everybody else is doing chide GB. That's why I'm must do it.
No, no. Really understand what the real value is for our customers. If you don't see any value don't do it.
But and if you see these two as a good ideas accordingly. You don't know, you know, the interesting thing here is I think we're we're at the beginning of the beginning of this not the beginning or the end of the beginning. Yeah, so I think it's gonna be interesting to see there'll be more ideas.
Let's pivot a little bit. Let's see. Let's talk about This cubecon we started off talking about it.
What are you hearing from? The people that you Booth I saw the booth was pretty crowded. Yeah, so Look, two three years ago when we'd all go to kubecon.
We'd meet essentially developers. Yeah, right or open source enthusiasts, right which is which is great and we have a bunch of those people here, too. But I've met a bunch of large Enterprise platform leaders.
Yep, we're seeing platform Engineers racing more sres look. We used to call system administrators. Yep, but jobs people absolutely, you know, you go to a reinvent your expect these people to be that which makes sense, right?
But here, you know, the percentage is significantly higher than what I remember from Valencia, right and it's disagree again. I think that's it that is indicative of how important and kubernetes is becoming to Enterprises. It's not a toy anymore.
They're running real production applications on kubernetes and they're looking for better and better ways to optimize their environments so that they can have a better operational infrastructure, right? So and look under today. That's our customer.
Thank you, right? Yeah, like we don't sell to a developer, right? The developer is my customers customer.
So look, we have to get our ICP right that from engineering and their customers is the developers. We have to make both happy but developers not my customers. So so we we look for those customers.
We understand their expectations in platforms you understand their needs and if I may say so myself people the best product in the market to solve their problems. That's why we have so many, you know, I wouldn't expect you to say anything. else you're there CEO, of course you built the best product you have but you know again, Kudos to you and the team because it's one thing for you to believe it.
It's another thing. for people to vote with their pocketbooks and their feed and their fingers and so forth and they have I mean, I've watched Rafa come up over the last few years here. Yeah, I remember so it's fantastic anything else new into show announcements news anything.
Absolutely so One big problem that we've thought about for a long time, but you know, we weren't sort of you got to be tall enough to solve certain problems, right? So we we announced and we are doing sort of a preview at the booth of a tool called environment manager. The idea is that a modern application is not just kubernetes.
It's kubernetes. It's Lambda. It's RDS.
It's Kinesis is all these other things that come together to, you know, build up more navigation. How do you orchestrate them all in a repeatable fashion? So we find that this is a very interesting Gap.
Right. So now terraform exists are from cloud exists and space Loft exists and n0 exists. And these are obviously very good products, but the but the sort of the the templarization and the automation needed like, you know do this for us do this next then try this thing.
Make sure this worked file a ticket do this next right kind of build a whole workflow of an environment. That seems like a gap in this industry. And we built a platform to do that.
Now. This is a early version of this. Right?
So we're showing you to developers right now just to get their feedback and we've designed it to kind of work very nicely with n zero or space left on all these products but this to us is the next sort of sort of you know, iteration of Rafi. We've we feel that on the kubernetes side, you know, we have a really strong foundation and it's gonna allow us to be a defensible platform for a very long time, but then we gotta go beyond kubernetes and solve the broader problem. Now, we're going to be working on this for two years, right?
It's not it's not it's gonna it's gonna take a long time to perfect it. But this is this allows Rafa to a solve the bigger problem for my customers same my CP same people. Yeah, but but the problem is so much bigger and it allows us to not be a kubernetes company anymore.
It allows us to be a modern application automation company. I think that is a public company. You heard it here first.
two years we get there. I hope so. Congratulations, it's good to see yeah good to see as well.
Thank you. All right. Did we mention the website?
Yes, right. I got CEO. Yes, sir.
I just want to make sure we tell people out there. All right, we're gonna take a break. We're live here at UConn.
We'll be right back.





