Brendan: Hello, and welcome to PodRocket. I'm Brendan. I'm on the engineering team here at LogRocket, and joining me today is Brandon Bayer, CEO and co-founder of Flightcontrol. Brandon joined us way back at the beginning of 2021 to talk about Blitz.js, and now he's back fresh from Y Combinator's Winter Batch to tell us about his new startup. Brandon, welcome. It's great to have you. Thanks for coming back for round two with PodRocket. Brandon Bayer: Thank you. It's great to be back. Brendan: So really exciting times for you. A lot of things going on that we could talk about from product design to fundraising, but let's just start by talking a little bit about Flightcontrol. What is it? What problems are you solving, and why should I want this product in my life? Brandon Bayer: Sure. Flightcontrol makes it easy for developers to deploy scalable, secure applications without any DevOps. You've probably heard about Heroku, things like that, so essentially we provide a Heroku-like developer experience, but directly on your own AWS account. The big problem we solve is the scalability and limitations of things like Heroku and just the limitations around configuration and stuff that a lot of these platforms like Heroku are just too simple and restrict you too much. Brendan: Yeah. I'm really curious about how you came to this idea. How did you get to feeling like there was enough of a niche or enough of a need from developers to be able to thread that needle between maybe the simplicity of a platform like Roku and the full-on flexibility, but also the complexity of just managing your own AWS resources? Was that something you found through your own work, through working with other developers? Where did that all come from? Brandon Bayer: So I had experienced the problem myself personally, but the real motivator was since I previously created Blitz.js, I noticed that a lot of Blitz users had the same problems. I come from this background, there's this spectrum in development, all across development. There's a spectrum between making something really easy to use, but highly abstracted versus a low-level abstraction, full power and control, but most of the time that's difficult to use. I started out on this journey on the side of I just wanted it to be really easy to use and I didn't want to think about CPU and memory and all this stuff. I just want to click a button and just have my app deployed, but I ran into the problems of that approach. Brandon Bayer: It's like no matter how much you want to just ignore things like CPU and memory for your app, you have to know about it. Unless you have a tiny side project where nothing's actually happening, but as soon as you get real users, you just can't get away from that stuff. Even if it's serverless functions or whatever, there's just a fundamental limit to what you can abstract, and you have to understand a little bit of that. So it's like, "Okay, this approach of making things too simple is not great, and so we need to try to find a balance between making it simple and easy to use, but also letting you have full power and control." Brendan: I feel like as you were answering that question and mentioning Blitz.js, I was thinking, "Oh yeah, this sounds like a framework developer talking about a problem and simplicity versus customizability." I guess it seems like you've spent now quite a bit of your career building tools and frameworks for other developers to use. Is that something you set out to do intentionally or what's drawn you to that work? Brandon Bayer: No. It was a surprise to me. So I've always been an entrepreneur. I always liked building things, working side businesses and stuff. I always wanted to run my own company full-time, my own software company. I was doing consulting for a while, because I didn't really have a great idea. But then I had this problem with JavaScript where I missed the Ruby on Rails type of developer experience in JavaScript, so that's what led to me starting Blitz.js, and that just has took off. I really tapped into a bunch of pin-up demand for that. It was through that I realized that, "Oh, wow. Not only do I really like building developer tools, I have a really good knack for it." So I was like, "Well, all right, I guess this is my niche from now on." Brendan: I'm interested in having a knack for developer tooling. What are the things you feel like that you do or that you're interested in that make that kind of work a good fit for your personality or your skills? Brandon Bayer: That's a great question. I think one thing is I just have a very low tolerance for unnecessary complexity, for inefficiency, for just stuff. If there's something that I shouldn't have to do, I don't want to do it. So there's that I really tuned into the pain and the frustration, and I have all these ideas, "It doesn't have to be this hard. It doesn't have to be this difficult. It can be so much better," so I have that motivation side, but then also, it seems I'm really good at simplifying complex things. This is really hard to do, but a lot of times the solutions, they just flow out of me. It's not a huge amount of effort for me to come up with a simple solution. Brendan: Mm-hmm (affirmative). Circling back to Flightcontrol, who do you feel like are the users that you want to find for this platform initially? Who are the people who are going to get value from this tool? Is it freelancers? Is it web shops? Is it small startups? Who do you see as the people who are best positioned to take advantage of what you're building? Brandon Bayer: All of the above, and to- Brendan: Good answer. Everyone. Brandon Bayer: ... add some more color there is if you are using Heroku or render or railway, then yes, we're absolutely, we believe, a better fit for you for now and in the future as you scale your company, but especially as you scale. Really, anyone that's running a web app or deploying a static site, we can handle all of those. You don't have to know anything about AWS in order to use Flightcontrol, so it's super easy to use, but since you bring your own AWS account, then you have the ability to pop the hood and leverage the real. Power of AWS should you need to. Maybe you end up hiring a DevOps person somewhere sometime to do more advanced stuff around security scaling, but you have the option, which is something you do not have with the other platforms. Brendan: Mm-hmm (affirmative). As somebody who's done a little bit of DevOps stuff in the past, my initial reaction of it is, "Oh, God. What happens when somebody goes and re-scales one of the machines or turns it off and turns it on again?" How do you manage the complexity of users accessing resources through Flightcontrol, but also having access to the platform themselves and being able to go in and press buttons and change the state of the resources they're using? Brandon Bayer: So the core powerhouse behind our system is Temporal. It's a technology that came out of Uber for managing complex backend workflows. It brings the simplicity of you know how async/await simplified promises and callbacks and all that type of stuff, so Temporal does the same type of thing to backend workflows. Instead of manually managing queues and scheduled jobs, Temporal lets you just write your code like normal synchronous code, but it's actually async running on multiple machines over multiple days, whatever you need. It's crazy the amount of complexity. Okay, let me put it this way. There's so many things in our background processes that I'm just like, it seems like it'd be so likely to have bugs, because of conflicts and state getting mismatched and stuff, but Temporal just makes it work. There's no bugs. It's crazy. Brendan: That's really cool. Was that a technology that you had worked with previously? Did you find that as part of your journey into building this product? Brandon Bayer: It was not something we used previously, but I knew about it and so I'm like, "Okay, we should probably use this, because we're going to need some help here." So it was a open source technology out of Uber and now Temporal is the company around it. I would highly recommend it, but it is heavy weight, so it's like a tank, it's like driving a tank. There's a learning curve. It's heavy. It's not especially nimble, but the power in the robustness you get from it is unparalleled. Brendan: I guess another thing I'm curious about is the platform as a service past landscape in general. You mentioned Heroku one of the OG players, Amazon, within AWS, there's three or four different past products. I feel like the question I always ask when I see a new startup in this space is, why does the world need another platform? What makes Flightcontrol unique beyond simply being more customizable or more flexible than maybe another offering ? Brandon Bayer: Though Heroku was created, I think, it's 15 years ago, which is wild to think about. They really, as far as I know either innovated, created this, invented this thing where you fully abstract AWS, and it's just a very simple service to use for deploying applications. Previously, that wasn't the case, and so they just made deploying things so much more to so many people. Over the last five, 10 years developer experience has become so much more important, and so there's pressure on these platforms to make a actual better developer experience, but also applications and things I've gotten a lot more complex, and so you need to do more things. Then there's also the scaling part. There's so many companies like Instacart, for example, that started on Heroku, but to reach a point-of-scale at which Heroku breaks, and so they have to migrate out to AWS. Brandon Bayer: It's like this just happens over and over and over. So it's like, "Wait a minute. Why is this keep happening?" Nobody has solved this. Everybody tries to create the next Heroku. There's been so many companies trying to be the next Heroku, but no one has achieved it yet. The big thing is because most of them try to fully abstract AWS. So you still have that fundamental scaling limitation because Heroku has a custom layer of infrastructure between you and AWS. So there's an extra lossy layer there, but with Flightcontrol, it's just pure AWS and so we solve that. Not only do we solve the scaling thing, it's also really easy to use, which is very difficult to do. Brandon Bayer: So we have people telling us that we are 10 times easier to use than one of our closest competitors. We had people saying, Yeah, we don't have to know anything about AWS," but as you use Flightcontrol, you start to learn more about us and so they like that because we surface some of the underlying stuff. Another thing is people end up needing to use some service from AWS, whether it's for hosting files or maybe they want to use Dynamo DB or something eventually. Then they have an app deployed somewhere else and then some other thing on AWS, and so when they look at Flightcontrol, they're like, "Wow, I can get the same exact developer experience that I have on this other platform, but directly on AWS, and then everything is all together, and I can use my free AWS credits?" Total no- brainer. Brendan: As I'm thinking about what else has happened in the past 15 years, it also seems like with the rise of DevOps as a concept, there's all lot more engineers who are primarily engineers, primarily focused on writing code and building applications, but who are at least a little bit comfortable with infrastructure and a little bit comfortable deploying code rolling back. So it seems like there may be a whole group of engineers who just didn't exist 15 years or ago who want to take advantage of something like Flightcontrol where it doesn't get in their way, but you can be a little bit closer to the underlying toolkit and the underlying platform. Brandon Bayer: Yep. For sure. One thing that we do really well at is we have an infrastructure as code feature that is designed for normal app developers not for DevOps. So there's things like Terraform and Pulumi that's infrastructure code designed for DevOps so you have to know AWS and DevOps, it's very hard to use. Our infrastructure as code is a simple JSON file that allows you to define your services, so you can have multiple environments and then multiple services within each environment, so a database, a web server, a static site, eventually I will be adding Reddit and background processing, and so it's really easy to just use and configure those for any normal app developer. You can do so many powerful things around service dependencies and be like, "Hey, I need my server to deploy before my Next.js front end," actually because the Next.js build has static props, and so it needs to fetch the data at build time, so we solve things like that too. Brendan: Yeah. Is that based on any existing infrastructures code tool or technology, or are you guys building that from scratch? Brandon Bayer: From scratch, yeah. Brendan: Interesting. Was that something you decided to do because you felt like just none of the existing technologies were the right abstraction or speaking the right language for developers? Brandon Bayer: Yeah. Yeah, definitely. There's nothing that is at this level of abstraction and it's just configuration on a file, so it's not like we're like making some new programming language or whatever, it's just JSON and config. Brendan: Yeah. You avoid all of the problems that you have with something like Terraform of managing state files. There's a bunch of complexity where it's powerful, but then you have to learn all this specific domain knowledge to use it safely and effectively. Brandon Bayer: Right. Brendan: I thought we might also talk a little bit about some of your startup journey and experience at Y Combinator. Since you guys, I think, just within the last couple of weeks came out of the Winter Batch, I guess maybe a good place to start, I'm curious, how you ended up at YC and what that experience is like in 2022? It's very different now that there's the pandemic, and I think it largely happens remotely. So what's that experience like? Brandon Bayer: So we decided to apply to YC last fall. We were trying to raise a funding round, but it wasn't going very well, and we're like, "Well, okay, let's just apply to YC. See what happens." I didn't expect much, but then I met a YC alumni that was real excited about what we were doing. They went in and recommended us, flagged us as somebody that YC should talk to, interview us. Then, we on Monday morning, it must have been last fall, got an email inviting us to interview you on Wednesday. We were like, "Oh, wow. We're actually interviewing, this is amazing," because I had applied twice previously with other ideas in the past years. So it's like you just expect to apply and not hear anything back. We interviewed and we got in and we were super excited and the actual batch was January to March this year, and it is just so much more than I even expected. Brandon Bayer: I just have so much respect for what YC has built. The advice that you get, the one-on- one mentorship with your partners and the custom-tailored advice for your company, all of the batch events, listening to these battle stories from other successful founders from Stripe and Airbnb, and then just how they totally flipped the cat and mouse game on its head with investors. Now, that you're in YC, suddenly investors are chasing after you. Then they have this demo day at the end where it's like, "All right, everybody is fundraising at the end. Everybody is talking to all these investors," and so you have so much leverage as a founder because the only thing investors are scared about is other investors. So when they know you're talking to a bunch of other investors, you can just do fundraising so quickly, because a typical fundraising process will drag out for a month or two; but with YC, it can happen in just a few weeks. Brendan: I guess I'm curious also, I know YC is very deliberately a short, three-month, incredibly intensive period where you're growing and meeting investors and working on your product. I guess one, how do you just stay sane during that intense period, and two how did you and your co-founder decide who was going to do what and focus on what parts of the product, the fundraising, the growth story? Brandon Bayer: So second question first, we were both doing everything essentially. I guess, overall, so I'm the CEO. I'm better at the product design marketing and sales. My co-founder, Mina, is the one that's building out all the AWS stuff. He's the DevOps guy. AWS still terrifies me a bit. I think it terrifies my co-founder too. Brendan: A healthy mix of respected fear. Brandon Bayer: Yes. Yes. For the most part, he's just building out the backend stuff. He's also helping do onboarding and support calls, but we're basically just, it's like you just do what you have to do. You just stay focused and build. As for how do we stay sane? It's like, something I learned early on with Blitz is I was forced to learn how to manage my own psychology. It's really tough. In the early days of Blitz, Blitz is, but then Redwood gets announced and I'm like, "Oh, no. Redwood's just going to totally destroy us." You have all these emotions about competitors and in market and things. You just have to get really focused on your inner confidence about what you're doing, and where you're going. It doesn't matter what other people are doing. Just because there's a competitor in the same space doing similar to what you're doing, the bar code is huge. Everybody doesn't want the same thing, and so now we have a great relationship with Redwood. Brandon Bayer: It's great because they have GraphQL, so it's like if you want GraphQL, then use Redwood; if you don't, then Blitz. So it's like, everybody gets what they want. We had to stay, and we still do as a startup, very intentional on a daily and weekly basis around, "How are my stress levels doing? If I'm stressed, stress comes from whenever you want something to be different than it actually is. So to resolve the stress, it's just a mind thing. All you have to do is just accept the current reality as it is, to truly accept it like, "Yeah, this is not the in state I want to be in, but I accept that today is how it is. I'm not growing as fast as I want, or I'm not being able to ship as much features as I want to ship, but it's okay. I just have to accept it," and it will change and do what you can to make a change, but you have to accept that reality. So it's like this duality is a super big thing in my life. It's like two different things that seem to be opposites, but you have to accept them both at the same time. Brendan: Do you feel like that's a consistent theme from other startups you talked to and worked with? Is that just the experience of being a founder? Do you feel like it's very different from person to person? Brandon Bayer: The stressors and stuff are very common. I think most people do not manage the stress of a startup very well. A lot of people end up burning out. A lot of people go too hard on it, and just don't manage the stress. I think it's super important to really get a good grasp on your own stress. That's how you're going to win over the long term. Brendan: Yeah. I think something that honestly comes up nearly every time we talk to people doing open source or in early stage companies is managing burnout and managing your stress level and being able to something over the long term, not just for a few weeks at a time. Do you have any other thoughts or tips or things you would want to pass on to people who might be in that same boat a few months from now, and what you would recommend for how they manage that stress and that experience? Brandon Bayer: I think it all comes down to self-awareness. you have to be extremely self-aware to know yourself, because burnout comes in so many different ways to different people. It's not the same thing. So you just have to be super self aware about like, "How are you doing? Are you getting burned out? How are your energy levels?" For me, for myself, it's whenever I start not being excited about work and it's like, "Oh, this is a drag." It's like, "Okay. I need to let off the gas a little bit. I need to go do something else totally not work-related." Brandon Bayer: Overall, I strive really hard to have a full day off work every week. That's almost a non-negotiable for me, unless it's sometimes I need to put in a couple hours on that day if there's some big deadline or something, but for the most part, I have a whole day off work. I make sure to do working out four to six times a week. I do rock climbing. I'm a pilot so I fly. So I have these things that I really love to do outside of work, so I make sure that I keep those in my life. Then overall, my philosophy is work hard, play hard. Brendan: Another thing I think is maybe interesting as a fellow podcaster, you have been documenting and chronicling your time at YC in your own podcast. I'm curious, what's the story behind that. Was that something you always had planned to do? Was that something you did at the spur of the moment? How did you end up doing a YC podcast? Brandon Bayer: So I started in the bootstrapping world and there's quite a few podcasts in that world from founders and co-founders, a couple top ones are Art of Product and Build your SaaS. I always found those super helpful and very inspiring, and I just love listening to them. I always wanted to have my own podcast like that, but I never had a co-founder until Flightcontrol. So then it's like at the beginning, we're like, "Yeah." He also have listened to these other podcasts as well, and so we're like, "Yeah, let's just do it," and it's been super helpful. We don't have that many listeners, but it's great for us just to talk about it. It's great to have it documented, and now if we back six months and listen to ourselves, it's so funny because you totally forget where your mind was at, and so it's great to listen. Brendan: Yeah. I'm curious, how much you feel like the podcast is for other people and how much you feel like it's for yourself, whether that's just to produce an artifact of where your head was when you were recording it or I guess to just have a forum to work through all the things you're thinking about with your co-founder? Do you feel like it's- Brandon Bayer: I think a pretty mix of all of them, yeah. Brendan: Yeah. Brandon Bayer: There's not necessarily one that stands out. Brendan: Cool. Then I guess my other question about that is, do you intend to carry this podcast forward? Is this going to be a part of your journey with Flightcontrol from now on, or are you planning to give it a pause and see what happens? Brandon Bayer: Currently, we have no plans to stop it or pause it, so we'll see. We'll see how long it goes. Brendan: Yeah. I'm curious, is this something that you've heard? Have any of your users found you through the podcast or has it been like a marketing channel for you? Brandon Bayer: I don't think anyone has found us through the podcast yet. I think more people find us and then find the podcast- Brendan: Mm-hmm (affirmative). Brandon Bayer: But we don't have enough listeners that downloads yet to be a marketing channel, but for sure, it could grow into that. Brendan: I guess last YC question, if you had to pick like one highlight, one low light, what are your takeaways from your time doing the YC batch? Brandon Bayer: I would say the highlight is just getting in, just getting accepted to YC, was just such an amazing feeling. Only 2% of the startups who apply get in, and so that was just a really good feeling. As far as low light, I can't really think of anything. I would say one regret would be not networking enough before the batch started with my other batch mates. The networking with them is a bit tough right now. It's all fully remote. I think it sounds like starting with the next batch, there's going to be some more in-person components. It'll still be mostly remote. So there's a lot of my batch mates, there's only a few of them that I've actually got to know. So after the batch, you continue to build relationships with them, but I wish I would've done that a bit more. Brendan: Another thing maybe to check in on with you is Blitz.js. I know there's still some stuff going on there. I'm curious, if you want to give us an update on what's happened since the law last time we've talked about it and what's coming up next? Brandon Bayer: Sure. So I think the last time we talked, it was probably before we actually made the fork of Next.js. Our original, how we had built it, we had a custom compiler that compiled your Blitz code into a Next.js code base. We reached the limits of that. And we decided to fork Next.js and move the log directly into the Next.js. That ended up killing our momentum, and we just stagnated our growth last year. We just, got bogged down, keeping up the data with Next.js and various headaches around the fork and stuff. It just as not working. Brandon Bayer: So in December we decided to pivot Blitz from an all-in-one framework to be a framework agnostic toolkit that you can add on to Next.js, any Next.js app or any JavaScript application. But we're starting for the Next.js highly tailoring that experience. The end result is that you will still get the same developer experience as you had before, but you will be interfacing with Next.js directly, and you just have the Blitz, like the Blitz authentication, the Blitz zero API layer added on top, but still the same developer experience. So we are almost ready to ship the first version of that. The initial version should be shipped by April 18th; that's because I will be giving a talk on it at React Miami, and so we have to have something ready, but it is going good. Brendan: Yeah. Great way to set a deadline for yourself and ensure you get something done. I'm curious, what kind of feedback have you got on that pivot from people in the Blitz community? Are people excited for this? Are people apprehensive about it? What are people thinking about this approach? Brandon Bayer: Overall? People are very excited. One thing that was amazing to me was one person was really pushing back. They were upset. Like, "Oh no, like I'm going to have to learn Next.js and stuff." I was like, "No, the developing experience are going to be very much the same." After I explained it more, they were like, "Oh, okay." I found out that Blitz was the first thing that they had ever used, like programming, they learned to code with Blitz. So they didn't know what Next.js is, or how similar Blitz is to Next. So that was amazing to know that people have learned the code using Blitz. A lot of people are excited about it. There's a lot of people who like Blitz, but have not used it yet. With the Blitz toolkit now, they're like, "Okay, yeah. Now I could actually use this, because I don't have to buy all into a fork of Next.js." Brendan: Mm-hmm (affirmative). Awesome. How are you envisioning balancing Flightcontrol and Blitz.js going forward? Brandon Bayer: So currently, Blitz is our distribution channel for Flightcontrol. A lot of our initial Flightcontrol customers are all, because they know about Blitz, so we'll continue to invest in Blitz. Right now, we have two maintainers. Flightcontrol is paying for two maintainers to work on Blitz. Some of them are helping on Flightcontrol as well, but we'll continue to make Blitz really great. We want to succeed over the long term. I think there could be a chance that we will spin out into the JavaScript Foundation or something like that in the future. But we're long-term committed to making it successful and seeing it become all that it can be. Brendan: Yeah. That's exciting to hear exciting and excited to see the next evolution of Blitz. Really excited to hear about Flightcontrol. Brandon, thank you so much for joining us today. Is there anything you would like to point our listeners to? Brandon Bayer: Well, you can follow me on Twitter, @flybayer. You can link to my podcast. It's called Flight Review, Flight Review.FM and Flightcontrol.dev is there. I'd love to have you try out Flightcontrol. If you have any issues, let me know, and I'm more than happy to manually onboard you to Flightcontrol as well. If you're already deploying somewhere and you want some help migrating your production application over, let me know. We'll be more than happy to help. Brendan: Thanks so much. We'll see you online. Brandon Bayer: Thank you. Speaker 3: Thanks for listening to PodRocket. You can find us @PodRocketPod on Twitter, and don't forget to subscribe, rate and review on Apple Podcasts. Thanks.