 Welcome to Amsterdam and KubeCon CloudNativeCon 2023. Join John Furrier, Savannah Peterson, Rob Streche, and UPSCOT as the Kube covers the largest conference on Kubernetes, CloudNative, and open source technologies together with developers, engineers, and IT leaders from around the globe. Live coverage of KubeCon CloudNativeCon 2023 is made possible by the support of Red Hat, the CNCF, and its ecosystem partners. Good afternoon, fabulous Kube community, and welcome back to Amsterdam. We are at KubeCon CloudNativeCon EU and it has been a jam-packed week. We've got over 10,000 attendees cruising around. There's 190 projects we've been talking about. There's startups to my right, there's cool companies to my left, and most importantly, John Furrier and I have some very fabulous guests. Rishi and Johnny, welcome to the show. You guys look great for day three. Johnny, how you holding up? Doing good. A lot of conversations, but all good ones. That's great. That's good. Fairly, I've had a few conversations. Rishi, how you holding that? Amazing show for us, as I was telling that for us, it's not the issue of the pipeline but of the wait list, right? So, signed up a lot, a lot of customers here, right? Good. All right. Well, how are you holding up? You sound like, this is our last interview. This, it makes me a little sad, a little happy. I think my vocal cords have maybe had their moment, but enthusiastically and intellectually, I feel great. It's been so fun. Really been getting a lot of energy off of our great guests, as well as the show floor. It's been lovely. And our team. Shout out to the team. Great. This is a great way to end it. Johnny, Dallas, 21-year-old entrepreneur, first company in 14, sold to Amazon. Youngest engineer working at Amazon, congratulations. And now you're at the kicking age, and you're a third startup. Rishi, I know we've known each other for years, going back to the big data world. You guys encapsulate, I think, the innovation. You got experienced system thinkers and innovators and the young guns moving up and taking names and kicking ass. You've been doing platform engineering since 14, when it was different kind of platform engineering. And that's the big story here. And I want to get your thoughts on it immediately, is it's now mainstream, but when you were doing it, it was like, you had to write your glue layer, write your cues. It was hardcore. Yeah. I call it prehistoric platform engineering now. We built a first platform, I was 15, 16, at our previous startup, pre-Kubernetes, built everything from scratch, Terraform, some Python, some Jenkins. Well, that's a heavy lift. Good on you. Yeah. That was a fun one. But it worked. I mean, one infrastructure engineer is able to scale 15 developers. Every single AWS region, thousands of concurrent EC2 instances. And now it's awesome to see the category kind of catch up. Platform engineering is a term now. It exists. We've talked about it in almost every panel this week. I'm sure it's been a lot of my conversations as well. What is the new company doing? Give us a quick milestone. I've got 50,000 developers using your stuff. What are you guys doing right now? What's the core product? What's the service? Yeah. So after building that initial platform, what we've done with ZEED is we really want to bring the ideals of platform engineering to teams that can't quite afford to build it themselves. I think everybody loves the idea of the passes, you've got Netlify, you've got Vercel, you've got all these really simple platform as a service companies that make developer experience very simple. But those don't really scale to actual enterprise companies. And so what we're doing with ZEED is that you have that developer experience on top of your own cloud, your own infrastructure, all in one, and then you can pick out pieces as you go to make it your own. Rishi, what's going on with you guys now? You got traction. I remember a couple of years ago, you were scratching this itch. AI comes around the corner. You're right there. Yeah. So for us it was pretty interesting because right before the whole GPT conversation came right. So we were doing end-to-end integration under that whole testing piece using behavioral environments. So there was only one missing piece that we were not generating test cases because it was impossible to do that before JNAI came. It was almost like the element 43 in the periodic table, which was missing for over 100 years. And the moment it came, it was right fit for us. So we jumped on it. Now we have the full support for all three models. I mean the OpenAI, the GPT-4 version. Then we are working very closely with Google Cloud. And then for the customers who are most paranoid, we are curating the open-source models. John, do you remember them? When you say paranoid, the ones that don't want to give away their information, is that what you're referring to? Yes. I mean, something happened with a very large company. Samsung. Samsung. That's been well reported. Yeah. And the stuff going on here, do you know who it is? We know who they are. Yeah. So a large company, they are very paranoid about it. So they are very eager, but they want the model to be on-prem, and they want it to be secure. So that's the piece we are working on. Right? So amazing traction there. Right? But we are working on all three, right? And just real quick, just end-to-end testing, using AI as the core piece of it? Yes. So test generation, test execution, all the way from the user story, all the way to the acceptance testing. Right? That's awesome. So we passed by it a little quick, but I want to make sure the audience gets to hear about this a little bit more. You mentioned starting businesses in 15, 16. How old were you at that point? So I think I joined my first start-up when I was about 14 and came the director- About 14. Just to repeat that for everyone. Yeah. And then entered high school, became the director of operations there and ran all the infrastructure, then on platform engineering, then our company got acquired when I was 17. So four years ago now, it's been a little while, but yeah. Got the cliff there. We're really into inspiring the next generation. We had Cassandra, who teaches the kids' day on our show yesterday. I'm really curious because I love to think of a young, ambitious person sitting at home right now, maybe watching this. How did you get the confidence and decide what you were going to build at 14? Honestly, I think the beautiful thing about software is it's all online and no one knows who you are. I don't know if you've seen that meme of a- Nobody knows you're a dog on the Internet? I got you, Johnny. That exact same thing applies to people. I knew very early that I love creating things. I love innovating, and I love helping people do more as well. I also knew that nobody would really respect me as a 14-year-old. It's not like I can go be a CEO quite like that, but I can write good code, and if I write good code, that can get me accomplished. So I started with development, did that for a little while, caught to an age where people grew a beard, people listen to me now. He already helped in terms of the business perception. Yeah, I can see that. I love that. There's a line that I love. It's so good they can't ignore you. And then as a woman in tech, as a young person in tech, I'm sure we share certain experiences. Do you feel like everyone's respecting the hell out of you now, and rightfully so? Much more so. We've got ways more to go. Yeah. It's better when you start making more money than all the VCs will start with money at you, ten ways from Sunday. Or you don't need their money. Yeah, I'd say no. That's the best part. Yeah, I was going to say more. Why are you going to involve the VCs? Stand in line. Yeah, this guy goes to a company at 17, and I don't think we need any venture backing at this table. I'm my own harshest critic. I've got plenty more to work on them. I'm excited for it. What's the big challenge right now for you at your business right now? Give it a little color commentary on what you're going through right now. Where are you? What's some of the challenges and how you're facing them? Yeah, so we started Z about two years ago now, really focused on, like I said earlier, developer experience. Make it very simple, make it very clean, make it very easy. What we've been doing more and more recently is focusing on the more offside of DevOps, right? How do you give an infrastructure engineer really powerful controls to customize this experience, and suit it to their setup, their devs, their environment? Because no company is the same. Everyone might look similar at a high level, but you get into the details. Customization is really critical. So just last week, actually, we launched a new feature, Blueprints, which is the first big piece of this puzzle, which is allowing you to take any infrastructure as code module and import it into Z. Make it a reusable piece of infrastructure. So if you've got, say, this is how I want to do S3 buckets at my company, bring that in as a blueprint. And now any dev can say, I need a new S3 bucket, and it will always use your template, your interface, and anybody, even an intern, can spin it up that way. And soon the AI, that'll be AI enabled. So much more scalable, too. Exactly. I think that's really the core of what this idea of platform engineering is, is how do you scale expertise? We've had agile, we've had DevOps, we've had all these kind of trends, but you really need software to do this, especially as we have more and more people becoming full stack developers and bringing development. Distributed all across the world. Yeah. I mean, there's so many reasons for that. The complexity is expanding in so many different axes. You need just software. Exactly. There's something about ZX that's going in the back, too. Oh, gosh, yeah. Multiple clouds, I think, is a really big trend that we're on top of, especially as all the AI stuff is picking up. GPUs and GPU portos, massive problem, harder to chat more. You guys are both doing AI. You're doing AI in your plan. Your young AI is going to be a tailwind from sure you're going to lean into it. I'm sure that you will not run away from AI. I can tell by your look. So, guys, share with us your perspective, because we've been trying to frame where AI will land in this ecosystem. I mean, it's not clear directly. Obviously, security is one. Automation is not a foreign concept that infrastructure people. But it's infrastructure. You can't have hose and agents. But AI could be a game changer. So, how do you look at AI in this market? A lot of plumbing, a lot of under the hood, a lot of log files, a lot of machine data. What's your vision of AI? How do you guys see it impacting the good, bad, and the ugly? So, yeah, you've hit it right, the log files. In fact, one of the secret ingredients we have is to use log files iteratively. If you think about it, when you have test cases, it's not that they're going to pass first time, right? Maybe 80% test cases will pass first time, right? In that, then we analyze the logs using gen AI, and then it becomes 85%, 90%, right? Then slowly it gets to 100%. So, it's like debugging in software, right? So, it's not that software doesn't work first time, right? So, it's the same thing here, right? Johnny, how is platform engineering going to evolve? Because you can connect to dots. You can see some things. It's not always clear. What's your take? Great, since you're the OG on the platform engineering side. Into the game before it was mainstream. What's next? It's interesting to see. I really do think multiple clouds is becoming increasingly important here. I mean, the GPU one is one thing here, but more and more we're having specialization of clouds. You have niche cloud providers coming out. You have privacy regulations coming up in Europe. You need to be spread across multiple different locations, and the orchestration of multiple clouds can't happen without a platform on top of that. And so, I think that's a real tailwind there alongside AI. The whole AI world is very interesting. I think the generated infrastructure is kind of one idea that people are very scared of right now. I think it's kind of funny. I think we're kind of already doing it. I mean, we have co-pilot and we have infrastructure as code. We already have lots of Terraform being generated by AI. It's already kind of happening out there, and I feel like people aren't quite seeing it that way, which is surprising to me. Well, I just read on Stack Overflow, they banned chat GPT. That's news. Oh, I didn't see that. They don't want any code pollution out there. So, how do you deal with the good code versus the bad code? I mean, aren't you worried that some bad code is going to come in? I think the evaluation loop is important. Right now, it's a lot of humans, but I think as we get more and more ecosystem built up around AI tooling, we're going to have automated ways of testing these things. I don't know if you've seen any of the auto-GBT examples where you write some code, you run the code, it breaks, you pass the error back into the AI, and it just iterates, iterates, iterates until it succeeds. I imagine you guys are thinking about this a lot with your testing. You have the evaluation loop already in place. Yeah, by the way, we were the first enterprise sponsor of auto-GBT, so I'm a big fan of the project, and we are also planning to contribute there, right? So, yes, absolutely. And as I said, so in fact, the same iterative approach which is using for coding, I will also help them with the testing aspect. What do you think we're going to be saying when we see you guys at KubeCon in Chicago in the fall and we're freezing there? Rishi, let's start with you. Well, I would hope that most of the open-source part has just started, like we talked about auto-GBT right here. So I think by that time, there will be bunch of stable open-source models, right? John can attest to it 10 years back, almost every week, there was a new Hadoop distribution coming out, right? So now, what I'm hoping that every month at least, right? There will be a new LLM which is going to become open-source and then all the training is going to happen on that. I think that is the next biggest thing which is going to happen here, right? Yes, you will have Microsoft, like you will have Google and you'll have all of them. I think you're on to something now. Yeah, I definitely agree. Johnny, you have any thoughts? Yeah, it's definitely interesting to see more and more of these platforms coming out, right? Like I think AI is so new by the systems there. I think from our side, the things that we're talking about are going to have a couple of cloud partnerships hoping to be able to talk about by then that are in the works right now as well as- We're rooting for you. They're going to be exciting. As well as all the tools here, it's very interesting to see what we've really done with this Blueprints functionality. What we just came out with recently is more and more of these specific technologies, like in the AI space, vector databases like Reviate. How does that get deployed? That's kind of still an unanswered question. How do you really scale that? We're going to have a lot more specific technologies to talk about. What do you guys think of KubeCon this year? Obviously the numbers are up. It's great, great numbers. Back to steady state for Europe, they oversubscribed by 50% of this, but it's 5,000 and got 10, 2,000 and they're waiting less. 60% of attendees, first timers, so you got a lot of action. What do you think about the presentation? I know the papers went out in November, so not a lot of AI in the track because it kind of surged on us big time. What's your take? Go ahead. I'm happy to be one of those 60%. This is my first KubeCon. Been to play with everyone. It's been great. Very welcoming. And yeah, very busy, 10,000 people. It's awesome. We had a couple of engineers who wanted to make it and couldn't quite get tickets, so I'm going to get them in next time. But yeah, also first time in Amsterdam, haven't been here in a while. It's been an awesome week so far. What's been the highlight of Amsterdam for you? Honestly, all the water and bikes, I love it. I'm always in the city. There's cars everywhere. I can like walk around. The city has been beautiful. It really does have an energy. And having been in the last few KubeCons, I'm not kidding when I say I'm getting energy off the floor. Everyone's so excited and helpful. I mean, the open source community is known for being a bit helpful. But I think one of the, and this is a vibe thing, one of the things that I've noticed between doing two events in North America versus coming here is it's just so much more inclusive. It feels like a different sort of family. So I'm here, I'll be curious to get your take more in Chicago and see what you're thinking. How's your time been in Amsterdam? What's been your favorite thing? English is not a problem. I mean, so last time we were in Valencia where people had some challenges here. They're like, here, there's zero challenges in speaking English. It's a great point on the language barrier. Very convenient. The other thing is that especially regarding KubeCon, day two, I would say most of the conferences have 30 to 70% strength. Here it was almost like 100%. Day three, no one ever is there. Here day three is about 80, 90%. That's a really good point, Rishi. No, I was curious. And Instagram too, by the way. There's a lot of reason not to be here. Yeah, the reason that things don't start till 1030, I think, around here, this particular event, I think everyone's being a little bit generous. But it's, I mean, it's just been, I don't know, I feel like the city is very much welcomed. I'm curious to see if the same thing happens in Paris next year. It should be great. Paris should be fun. I mean, who doesn't love Paris in the springtime? Can't help myself. It's been a great show. I got to tell you, I love the show. I thought it was dynamic. We had great segments on the Kube. And I think it's more human this year. It feels more inclusive. And they've both been inclusive. And the partnerships have had an opportunity to mature. I think we had a lot of things happening in silos for a few years there. And now everyone's starting to team up and work together and really start to build some very critical stuff. Like I said earlier, when you see data protection on the agenda, you know, it's an enterprise-grade show. I appreciate that that's your benchmark, John. I'm going to have to figure out what mine is. Most of the cutting edge show, no, this is like a benchmark. It's more of an observation. Like, you know, the developer shows are, they're hardcore developer power, you know? And you got coding, clinics, all kinds of deep dives. And sometimes some of these other enterprise topics don't get addressed on those core tech. So when you start to see things like data backup and recovery, data protection, where security's in there, it's like, okay, it's hardening to be enterprise-grade. That's why I kind of like this platform engineering repositioning because platform engineering two years ago isn't what they're talking about right now. It's not, it's a completely different thing. It's more mainstream IT. And to the point of like inclusiveness and dev conferences versus cloud conferences, we're talking so much about developer experience. I mean, Backstage is the new CNCF project that everyone's talking about right now and a lot of focus on. It's really how do we take the stuff that we're talking about here at KubeCon, bring it to developers and help them be a part of the conversation. And I think that's the inclusivity as well. That's a good question. You're the first guest to bring up some of the trending topics. What are some of the other ones? Backstage, the developers of Jazz about, what are some of the other projects that are in the hallways of getting a lot of traction? What would you say? I mean, I think similarly, it's not necessarily as new but GitOps continues to be very interesting. The GitOps con for the first time this year, it's very similar premises. How do you give developers, the end users, the ability to control this stuff? Infrastructure as code, Git, these are the interfaces that developers naturally live in. Now we're pushing more and more of this to the edge while making sure that they don't have to deal with all the complexity of data management in the actual infrastructure. Let's keep that over here and give them control. Well, you're a back end guy so the web assembly is actually hot trend too. We had awesome guys on. Rishi, what are you seeing as the hot developer? What's the catnip for those developers? So, developers, I would segregate them into two categories, right? So, most of the developers are eager to learn, right? Because, I mean, there was a dream for people to become 10X developers, right? Now, you are either going to be a 100X developer or you're going to become irrelevant and most of those are ready, most of the developers are ready to become 100X developers, right? If you're not a 100X developer, you're irrelevant. You're irrelevant. It's a bold claim. Catnip. Yep, there you go. Dog bone, dog whistle. Are you going with the cat and dog in a yang? You referenced in the dog t-shirt from yesterday and I see what you're going, John, don't worry. What's the best swag you've seen? That's a good question. We do a swag segment, so yeah, no, I'm curious. Curious if you agree with me, no, I'm kidding. I don't know if it quite counts as swag but I saw one of the usage-based platforms handing out dollars, handing out free money. Saved cost-natable us. That one was pretty good, I got me. It's clever, it's catchy. What about you? Well, I would say I was swagged at tote bags, I mean, they were, we were out of them very soon, I mean, I think day two itself, right? So everybody was carrying them, but I didn't get a chance to go and check at other places. Well, we had a big debate. Do we like the Oculus Rift and the Legos that IBM had the higher, bigger price but not as creative as your favorite? The Replicated Apron. Apron, yes. The Apron and it's a nice one, it's a utility Apron for the barbecue or whatever, it's got a little leather straps, yeah. And the Hangover Kit, they have a nice little eye mask for the van amongst us, no, the whole thing was great. I got one last question for you guys as we wrap up this fabulous segment and I'm going to start with you, Johnny. What would you say, because you've obviously been a bit of a gateway to people into this space and I can imagine a lot of people look up to you. What would your advice be to someone who's just learning about Kubernetes, KubeCon, the open source community? What would you tell them? I mean, honestly, the space changes so fast, I would very much focus on whatever is brand, brand new. Be at the cutting edge because that's going to be the heart of it all in a few months to a few years. So AI, LASM, platform engineering, backstage, these are the things to focus on because that's going to be all of the conversation for the next several years. LASM is very exciting, I mean, even if we talk about the prehistoric platform that I built back in the day, we had to go to containers, that was a necessary step for us, we couldn't live in the supervisor room. There wasn't another option. And LASM just kind of improves that functionality, like we're going to have more and more of these back-end runtimes that are accessible to developers and usable by infrastructure and I think that's very interesting. We're going to have to start paying you because you pretty much summarized all the topics that we had here on the Kube this week. So I very much appreciate that. You should have put us out of business one second. I'm having a drink with Matt Butcher for me on after this too, so I'll let him know that you're excited about what they're doing. That's very exciting. Rishi, what about you? What are the questions? A question? Nobody knows what's going on anymore. We're off on after this. What would you say to someone who is just learning about the open source community, CNCF, KubeCon? What would you tell them? What would be your invitation or your advice to someone who's just getting ramped up? Yeah, so I would say that, I think we have seen Kubernetes evolving and now it has almost become like an infrastructure play, right? That something which I found most exciting about that all of the new innovation, even in the JANA AI, they are assuming it's going to happen on Kubernetes and containers, right? That that's kind of assumed, right? So I think that's the biggest thing that I mean. And do you think that's going to be real? Yeah. Okay, there it is. Well, Savannah, it's been great run this week. Rob Streche, we have Yop, great team. Yeah, we've had a really, really absolutely killer team. In fact, I want to give them all a shout out. We've got Yop and Rob as my co-host. Noah, thank you very much. We've got Steve. We've got John. We've got Leonard. We've got Rihanna and we've got Kony. None of this is possible without the fabulous production team that is seated over here to my right that you never get to see, but they're beautiful and they're brilliant, which is quite a bang. We are very lucky to be working with all of them. Rishi, Johnny, thank you so much for joining us on the show. John Furrier, thanks for making the magic happen for all of us. And thank you for tuning in to our live coverage all week from Amsterdam. We're at CubeCon Europe. My name is Savannah Peterson and you are watching the Cube, the leading source for emerging tech news.