 I have no prepared questions because this one is going to just be heartfelt and easy. And look, I don't know if I've ever had the chance to say thank you because that was really a course trajectory change for my career, seriously. So I just want to say thank you now. Thank you, Cathy. Thank you so much. So we're going to start with like Clow Kick. Before CoroS, you're a founder, and maybe what stage did you start this particular company and why not just go to work at a big tech company like Google, Microsoft, or Amazon? So tell people a little bit about the first startup you did, why you chose that route over something else, and how did that end up? Sure. So my career has been weirdly just connected. It's been a snowball that just continues to build up. When I was at Oregon State, Go Beves, I met this guy who is running the largest open source projects in the world's FTP servers. So it was when Apache and Mozilla and the Linux kernel and FreeNode were all running their infrastructure there. And I got a job my first year in college helping run those servers. And so that connected me to open source and servers at the same time. That led to working at Mozilla where I was on the add-ons marketplace at Mozilla. And at Mozilla, I met some guys that did this weird thing called Y Combinator. They were these Y Combinator founders, and this was like in 0607, and I had always wanted to go to Y Combinator. Brandon and I were, Brandon's my co-founder in CoroS, good friends from Oregon State. We applied to Y Combinator like 06, the very first time they ever even had a batch, and we got rejected. But we met some real-life Y Combinator founders, and we were like, let's go start a company. Tell people a little bit about what Y Combinator is. And I think some people are familiar with these startup incubators, but maybe give people a little bit about what Y Combinator is, and then why were you so interested in going to one of those incubators? Well, the thing that Y Combinator did differently, and YC really started the startup incubator concept. I mean, it was really the first one ever, and still remains kind of the premier one. But the thing that Paul Graham did that was unique is he really reached out to technical people, and was like, okay, if you're a technical founder, you can start a company, and we want to help technical founders start a company almost as an alternative to getting a PhD program or something like that. So that just really spoke to me as someone that loved building things and communities and software to be like, oh, I want to go start a company. And so when I met some actual people that were in YC, that was sort of the connection to Silicon Valley for me, coming from my Christmas tree farm in Oregon sort of thing. That's how I got connected to it there, and then I started Cloud Kick and that really snowball just started getting bigger from there. You know, you have this kind of origin story, like the things that drew me to open source early was this fact that people doing all this important work are so open to onboarding the next generation of people that are interested in this work. And I think you find that in your open source roots, pre GitHub, by the way, this is when you upload your source code to some FTP server, and you're responsible for a site telling people where that FTP server is. SVN, baby. Oh, SVN. Okay. Okay. There's some CVS people in the crowd. I'm an RCS guy personally. And then this idea of starting a tech company, now that I think about it, how would anyone know how to do that without a thing like Y Combinator in place, right? Like there's no blueprint for that. That's just the thing that if you know some people, you know some people. In this case, they're advertising that we're going to democratize this process for anyone that's interested and kind of give you almost like, feels like a drive run. Is that how it felt to you? It was almost like a college program almost. I mean, there was a cohort and a class and then they teach you how to pitch and all the stuff. I mean, we could write the software, but how do you talk to an investor? What is a cap table? How do you work with a lawyer? You know, all of these things. And probably the biggest blessing I've had in this whole thing is that I had CloudKick, my first company where I could go really screw it all up. You know, that was like where I learned all my initial lessons on all this stuff that you have to learn it somewhere. And that CloudKick turned out, you know, net well, but it was where I was able to double down on that with CoreOS and make a much bigger impact. One thing people struggle with is where do you get the idea from? If you're a maintainer of an open source project, I've done this, scratched my own itch, put it on GitHub, and then the star count goes up. And look, your idea probably has product market fit. You kind of know what you can do and you can make a decision to build a company around it. Was CloudKick something that you just said, hey, I'm going to start a company and then you have to figure out what to work on? Or was it a problem you already had that you just chose to solve via the company angle? CloudKick was very much a, oh, we're doing a startup. We have to figure out how to make a company. And within YC, which is three months, we pivoted four times into different things. And what we ended up on was back to what we knew, which was infrastructure stuff. But it was all over the map. CoreOS was the opposite. It was very much like, here is a very clear vision and here's what we want to go and achieve. But CloudKick, for sure, started out as the startup mess. Again, part of the lessons learned there is, one, pivoting is okay. But two, also, if you have an idea what you want to do up front, that's great as well. Talk about some of those pivots, right? Because I think there's rumors out or facts out that Docker starts out as a competitor to Heroku, right? They're building a pass for the next generation. And then happenstance, there's this Docker thing that falls out of it. And then the whole company pivots around it. And there's things like Slack comes from, like, I don't know, a gaming or some whole non-adjacent business. And then you end up with Slack with this multi-billion dollar company. When you start CloudKick, what were some of those pivots? Were you selling pinball machines to start with? Like, what were some of those initial ideas in some of the pivots? So first to set the stage, I was 23 years old. I had never done a startup or anything on my own. And I had no idea what I was doing. So just prefacing that. So the very first idea we applied to YC with and what we got accepted in was called Scholarshipster. And it was crowdfunded scholarships for students. Because I was a fresh college grad and that was my best idea of something that I do. Now again, I did not get to YC because of that idea. I got to YC because I knew people that were in my Combinator and they recommended me as a technical person to get into YC. But very quickly, I mean, talking in the matter of weeks, we pivoted the infrastructure software. I think we probably wrote like a landing page for that. And then it went away. Yeah, so I mean, it went from there to a like a cloud NAS sort of product that was always backed up to cloud sort of thing. It was like Dropbox, but for the home server kind of thing. And then we moved to Cloud Kick because there was this novel thing that had just been released called AW Web Services. And it had these like APIs that nobody really understood and it didn't have any control panels or anything. So we were like, why don't we make this really easy to use for people? Because it seems like it's going to be something and ended up building, you know, a control panel for it and tools to monitor AWS and and some other things. And that was right was AWS was was getting going. And it was just EC2 and S3. So Cloud Kick comes out, fills a gap and you did an IPO, Cloud Kick. You got acquired and looking back, what do you think the reason for the acquisition is? Is this something that you felt the acquiring company just needed? Do you feel like the runaway was running now and you just had to do something? What causes you to have to exit Cloud Kick when you did? I mean, again, I was 20 now 24 or whatever the Cloud Kick from start to finish was two years before we were acquired. And just I I was just trying to figure out when someone knocks on your door and is like, hey, we want to buy the company sort of listen. And again, remember, I grew up in the sticks in Oregon. And so like any level of money was very helpful at that point, too. You know, and so it was like, somebody shows up and it's like a life changing thing for me and my family and my, you know, my folks and everything. And so, you know, they, they racked space came along as they were trying to compete with AWS and was looking to build out both some more of their cloud product line and have a San Francisco presence and everything. And it just seemed really interesting. And then at the same time, there was other company, another company interested to and then it was kind of a game was afoot once we had two of these companies kind of chasing us to acquire us. There's this illusion in the start of space where a company gets acquired, people read that headline. And to them, you've made it, right? You can just go retired by the farm. But there's very real things that happen, right? There's a check, there's lawyers, there's paperwork, and there's this thing we call the golden handcuffs. This is the thing where they say, Hey, you don't get to just abandon chip. What did those golden handcuffs look like? Right? What are the next year? Like, you don't just sell it and walk away. You sell it in, do you become an employee? Do you become a lead product manager? What's day two look like after that check cash us? Like, what's the technical implementation of the handcuffs? Well, I mean, do you, do you just get a big check in the mail and you just go to your bank and deposit it? Like, or is there like a payout? Like, what do you do? Well, both it's startups financing and an acquisition. It is kind of a funny thing when you go to your ATM and then all of a sudden there's much money in there. Because even when you, even when you start a startup, like you get this big wire and it just goes to your checking account of your company. And you have like, you know, normally you at least have a debit card. So I remember checking for our financings that like CoreOS or Cloud Kick or whatever, by just going to the ATM is like, Hey, yeah, balance. Oh, yeah, you know, kind of thing. It's kind of crazy. And so then on the personal side, too, I mean, yeah, that, that kind of happens typically that there's some portion of that happens when you close. And then, you know, there's a couple of different mechanisms start up or company, the big companies will use to retain you. Often it's a some form of equity grant in the company. And in both of my deals, they were essentially holding back some of the portion of that I would receive if I had sold all my stock in that moment to the company for a later date sort of thing. So the reason why I asked that question is, you know, when a company does come knocking, now that you know what you know, you know, that's part of the deal. That you're going to probably now have to join that company and be a team member and including all the people that end up working for you. So let's, let's pivot to the CoreOS situation. You get bought by Rackspace. Well, Cloud Kick at Bob at Rackspace, CoreOS got bought by Red Hat. I know. Okay, I thought you said to the CoreOS situation. How about the right time? Maybe I'm a source. Got my C's confused and then my R's. Cloud Kick gets bought by Rackspace. You stick around, handcuffs fall off, and you're making a decision now. Do you do this again? And my guess is it's much easier to raise the second time during that time period. And did you have a clearer idea? What do you want to do? Or you just knew that you had to start another company. This is going to sound extremely Silicon Valley cliche. So please bear with me. But I was on a bit of a like, you know, personal journey quest thing. I was, there was no psychedelics involved at that point. But I was literally on the trans-Siberian railroad crossing through Russia, talking to a Russian horse meat salesman about how to secure the internet. And aha, I had this idea for CoreOS. And I did you not. That's how it came up. And at that point, I'm like, OK, I need to start a company around this. Let's do this. So I called up my buddy Brandon, pitched in the Indian. It was on. So your goal was to secure the internet. You didn't quite know how, because if I hear someone say they're going to secure the internet, I'm thinking you want to start. I can't put the idea there. Like the whole CoreOS thing. I mean, it all just came to me in a flash out the window somewhere in Mongolia. I doubt that psychedelics were not involved. Sometimes you don't need them. So you're in this mode now. And I when I hear secure the internet, I'm thinking network equipment, switches, firewalls, routers, things that sit at the edge of the internet, point of entry. And you said you had this vision. Was it CD, the operating system, the whole Docker containerization thing? Was that part of that original vision? I mean, the components we knew were going to need to be part of that CD, you know, Docker and all the stuff for like implementation details as we went and developed it. The whole key, and again, for the former CoreOS that are in the room here, they'll have heard this a million times. But the whole idea is like the way we can secure the internet is when there's a security vulnerability, all the bad guys write programs to go automatically hack everything on the internet. So what if we could write programs to automatically update everything on the internet? So what does it take in order for all the server infrastructure to be really easily updateable? Well, in order to do that, we need to run things as distributed systems so that any machine can go down at any time. And we know that some companies can do this well. Folks like Google, who have this thing called Borg, and they somehow figured it out, but at that point, nobody really knew how that worked. So we're like, let's go implement all the components so that any company can go build their own Borg-like thing. And so we first built an OS, the updated itself, and then we built XD, which is required for distributed locking, and then we started building a container runtime system. And then this weird thing happened. Google shipped Kubernetes, and they were using XD, our database, and they didn't talk to us about it or anything. And so we were, again, continuing our Silicon Valley cliche. We were sitting in a garage in Palo Alto, and Kubernetes ships this thing using our software. You think they would have at least asked us, you know? And then we're like, well, we better just move to that. And kind of that's part of how this all emerged, is we just always wanted to contribute the best in class software or fill in the gaps where they're needed. So that's why we built some things, but we also leveraged a bunch of other things, such as Docker and Kubernetes. Where I entered to this story, I'm working at a company called Puppet Labs. This is a startup that was small, that becomes big, and they wrote the wave of DevOps transformation. And I remember we would go to companies and talk about legacy software, broken promises, and say, hey, you got to get with them modern times. You got to disrupt this. And I remember being at CoreOS when Docker comes out. And I remember you could not walk around the company and say containers. We thought it was a fad. It was kind of against this whole declarative nature of how to do things. It was like this thing that came from the developers, not the system administrators. And I remember trying to do something in Golang. It was off limits. Everything had to be in Ruby, mainly because Golang did not support AIX of all things because they were targeting enterprise software. So the disruptors became a little timid to innovation themselves. And so I'm looking around and like, yo, this is not going to work for me. I got to go do something else. I'm into this Golang. I ended up taking a different job. I was at New Relic for a short stint. And I started writing things in Golang. And one of the first projects that I wrote was, leaving Puppet Labs, I did not want to do full-blown configuration management anymore. Number one, working at a company that uses configuration management, dog-fooding their own product, it gets insane. Instead of using OpenSSL to create SSL certificates, they write a Puppet wrapper over a thing that could have been a shell script. And it's just too dogmatic. And I was like, you know what, going forward, all I need is a file package service and CompD was born. And CompD is just like, all we need is the metadata. And that's how I found EtsyD. I was like, oh, what's this EtsyD thing? It's written in Golang. CompD will be written in Golang. And I open source CompD. And I remember going to Fosdum for the first time and someone was on stage doing a presentation on CompD. And I'm sitting in the back of the room like, how does this person know about CompD? Just released it as a hobby project and then it spread like wildfire. Then we meet at maybe GopherCon. GopherCon. And GopherCon, I'm sitting in the audience and I'm about to give a talk, Go for Sister Admins. And bless their hearts. Brian Kalestyn, Eric St. Martin, they start this conference for the community. Two developers just wanted to see a conference for the programming language they love. They were terrible MCs. It was not good. It was so bad that I was sitting in the audience and I was like, what backstage? I said, hey, you don't know me. I'm Kelsey. But can I try to be the MC? And they're like, sure, try your best. That whole time. How far you've come? I've come very far. It was actually pretty good. But the whole time, I think I was kind of in contact about being at CoreOS. We were trying to figure things out. And that whole conference felt like an interview. Because you're contributing to the open source stuff. They see you're building on top. You're kind of getting to know people. And then it was like, what do we do? They're like, you could do this. Do we need more of that? And I remember you just making a call like, let's just figure it out. In your mind, when you bring on someone like me and you say, what just figured out as a founder, you're responsible for every person. Because I remember coming to the office for the first time. I was like this very small thing. You used to call it the farm. And I was like, literally, this is the office. Is it safe? From a garage in Palo Alto to under a freeway in San Francisco. The Uber driver was like, this place? I was like, yes. Again, it's just so ridiculous. Below us, there was like a K through 8th school. And around the corner, there was a lesbian porn studio. I didn't know about that part. You did. Sorry, neighbors. Yeah, pink and white. So you make this decision to bring me on. Every employee you hire in the startup is a very calculated thing. Because you don't want to disrupt team dynamics. How do you make those hiring decisions? Especially when you don't know exactly what someone's going to bring to the table. Yeah, I mean, I think one of the things I appreciate most about Cora West is the team that we built. And just to see the inter-team camaraderie and everything too, even to this day. Like, it is pretty cool. And I think in terms of how do you hire folks when? I mean, that's all these startups are, is the team, you know? And so hiring based on just, you know, they call it, I don't know, culture fit or whatever. But I don't know. I think that there's just like this ethos that was a shared understanding that happens when you align around strong vision. And if people get into it, then they're into it. And if they don't, then they don't get it. And they maybe just shouldn't be part of the team sort of thing. So it's kind of almost like, are they willing to drink the Kool-Aid or not, you know? But I think we just hired really people that are really technical, that really believe in what we were working on. And that wanted to be part of something bigger than themselves. And, you know, that's one of the things that I feel like is prerequisite for anyone to work on a company, including myself. Like, I want to work on something bigger than myself. And I want to work with excellent people. And I want to work on really hard technical problems. And so by putting those things sort of out there at the forefront, it just brings the right people together almost naturally. Maybe it was after the evaluation phase, but I can remember early on we were trying to hire the first salesperson. Don't worry about the salesperson. But I remember where it was like three of us in that room might have been me, you, and Brandon Phillips. Okay. And he asked to explain the company what they did. And Brandon gave his explanation. You gave yours, which still sounds as insane of securing the whole internet. He was just like, secure the internet, huh? Okay. And I remember I gave this very methodical answer that I think made the light bulb go off for him. And I think I saw you look at me like, damn, that's a really good answer for what this is. And then there was like this trust unlock. And part of that trust unlock was, Kelsey's going to go explore things. And I remember one of those things being Kubernetes. I'm working at CoreOS. We got this fantastic vision. People doing a great job executing. We had went all in on system D that leads to Fleet, which was the orchestrator for CoreOS. And here I am like, hey, this Kubernetes thing, it's interesting because I remember how it came to us. We got a call from maybe Google team saying, hey, we're going to open source the thing. It's going to come out tomorrow. Here's your embargo. Good luck with that. And we were not necessarily in the red hat Google coalition at the time. And I remember the night before you, I, Brandon, Melissa, she's in the audience somewhere. We were like, what are we going to do? And it's like, hey, we got access to the GitHub repository. We should like, when they announced it, we should like, here's how you write it on CoreOS. And I remember that was the first start of like Kubernetes the hard way. Getting access, knowing that we can publish something by tomorrow because we only had like a day heads up. And all night we were building out this guide, how to run it. And we had Rob and he put it on the website, made it pretty, all the graphics in one night. And as a startup, I remember vividly, sometimes you only have six or seven hours to be included in the wave or sometimes get left behind. I'm pretty sure we could have made up ground. But I remember when Hacker News came out with this Kubernetes thing. And I remember being so proud that, and I think we all in the office figured out who had the most karma on Hacker News to do the post. So we had the best chance of getting a ranked tie. And CoreOS was number one. We put this guide out. If you wanna try this thing that Google just announced, here's a step by step guide on how to do that on CoreOS. And most people had never seen this thing before. There were no set up instructions in the early repository. And so that kind of thing, I really appreciate you trusting the individual team members to own their areas. I remember Mike owned the OS. You trust them. He wanted to start doing some rust thing. You were like, sure. And I think as a founder, what allows you to give that trust knowing that this is your baby and you had a co-founder, but you're ultimately responsible if this thing works or not? We're in Europe. So I'm going to try to use a little bit of German, which I don't really speak well, but there's these emotions that only have German words. And one is called Schadenfreude, which is like getting pleasure from other people's failure. And so that's like watching trainwreck or something. But then there's the other one, which is Freudenfreude, which is getting pleasure from other people's success. And I think a lot of what motivates these things is Freudenfreude in terms of like, let them go for it and just see if it works and if it does, how amazing will that be? And so I just try to live as much Freudenfreude as I can. And I appreciate that because that was a big signal that I had something more than the core talents, the ability to let me go out there and discover who I was, especially on those big stages. And I remember once that Kubernetes thing happened, a couple of months later, I think we sat down the whole team and we were like, we're going to be a Kubernetes company. You made this pivot. I think you remember you asking me, hey, do you want to lead product? I think you're onto something here. You had the vision. You've been a good steward of where everything was. And I think you get it. Would you like to lead product? So I appreciate that opportunity and turned out well for all of us. We're wrapping up because we have time, but CoreOS does really well culturally for tech. It wasn't just that containers were hot. CoreOS gave people a usable way of using that technology and we kept filling gaps almost like weekly. A lot of people don't know. CNI comes from CoreOS, SCD comes from CoreOS. A lot of the things that people take for granted come from CoreOS. Looking back, you get acquired, that was successful. You did finally get to go retire, buy that farm, start flying airplanes. You went full circle. You look back at that and I think culturally, CoreOS remains with us all, even if the name is changed or lives on through other projects. Do you feel like that's it? As a founder, you've completed the puzzle of building something sustainable and now being able to walk away or is there another company in you? I just haven't had my moment like I did in Mongolia with that clear vision of seeing what we're doing next. But what led me to CoreOS was just this relentless tinkering with things and I continued to tinker and build things. And now I'm running my bare metal Kubernetes cluster and all that sort of stuff still and building and shipping stuff just on my own. But nothing's hit me like a lightning bolt like CoreOS did. And maybe if something else does, it maybe it will or will not be another tech company in the same form. So I don't know, we'll see, but I'm just gonna keep doing my thing and I'm confident it'll all work out. Alex, I wanna appreciate your contribution to this entire space. A lot of the cloud native ethos, the technology, the culture, the people. And you always have this one thing is all we have to do is keep going. And I think a lot of people who contribute to these projects understand that very well, regardless if it's a commercial company or not, all we have to do is keep going. And thank you for helping a lot of us get started. Thank you, Kelsey. Thank you.