Bolt.new CEO Eric Simons — one of the fastest growing AI companies in the world

Eric Simons is the CEO and co-founder of StackBlitz, the company behind Bolt.new, which reached $40 million ARR in five months after launching in October 2024. Bolt.new is an AI-powered platform that allows anyone to build, edit, and deploy full-stack web applications directly in the browser. In this episode of World of DaaS, Eric and Auren discuss: The explosive growth from $0 to $40M ARRPioneering usage-based pricing for AI toolsCompeting with Replit, Lovable, and VercelBuilding WebContaine...

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:02Auren Hoffman?

  Hello fellow data nerds. Welcome to World of DAS. I'm your host, Orin Hoffman, CEO of Incubate and GP of Flex Capital. Discover more episodes, get weekly data as a service news, original content, articles on data and more at worldofdas.com. That's world of das, D-A-A-S, dot com. Hello fellow data nerds. My guest today is Eric Simons. Eric is the CEO and co-founder of Stackblitz, a company behind bolt.news, which reached 40 million ARR in five months after launching last year. He previously founded Thinkster, which was acquired in 2018, and pioneered web containers technologies that enable the full node.js and development

• 0:44Auren Hoffman?

  environments to run entirely in web browsers. Eric, welcome to World of DAS.

• 0:49Speaker 01

  Awesome. Yeah, thanks for having me. Excited to be here.

• 0:51Auren Hoffman?

  And we're investors in Stackblitz, very happy investors. I feel like there was four companies that all kind of launched roughly around the same time with similar products. We had Stackblitz, which is the bolt.news, we had Lovable, we had Repli, we had Versal, all of which have very good products, different but very good products, all of which just went completely vertical immediately. There was complete product market fit. Walk me through that time.

• 1:19Speaker 01

  Versal, they had the first product called B0. It had been out for a couple of months by the time that we had launched, but it was very simple as far as the sort of applications you could build. It was really more almost of a purely designed tool at that time. You couldn't like build real working applications with the thing. And that was really what we saw the opportunity as was going and building full stack software as a non-developer that also looked great, etc. That's why we went and built Bolt. We launched it in October of last year. We were really the first vibe coding product that hit the market. And we just took off like wildfire. Two months after that is when Lovable launched and then Replit launched some of their agent

• 2:03Speaker 01

  stuff, three months after, four months after, something like that. Once we had launched, everyone saw the potential of having full stack AI agents.

• 2:13Auren Hoffman?

  Really helping build your application, essentially.

• 2:15Speaker 01

  Exactly.

• 2:16Auren Hoffman?

  And people were using Cursor before that, if you're a real developer, but not the same as building the full app.

• 2:23Speaker 01

  Well, we cracked the lid on by being browser based and not having to be an IDE with Bolt. It allowed product teams to really build software themselves for the first time. So instead of having to go harangue developers to ship prototypes or whatever, they could just do it themselves.

• 2:37Auren Hoffman?

  Today, that's what I see. Before the PM even goes to the developer, the PM is building the V1 or V0 themselves. They get it done in Bolt new or one of your competitors. And then they bring it to the engineer and they say, hey, make this better, harden this, make it better security, do all the other stuff that you need to do, but like here's the working prototype.

• 2:57Speaker 01

  Exactly. It's kind of crazy now because like a year ago, it was so new, that concept. Now that's just really the default of how a lot of companies are operating and I think everyone's going to. That was the big paradigm shift. We were really the first ones to kind of kick that off. And then there's been a number of different companies that have entered in the startup side, also incumbents, like Big Ma has their own thing that's kind of like this. Wix bought a company and they have their own kind of I'm putting thing.

• 3:22Auren Hoffman?

  And then you have, I'm doing Bolt new just for mobile apps. There's almost like little spin outs, things like, I'm going to do it just for this particular thing or something.

• 3:31Speaker 01

  Exactly. We do it for mobile apps too. So it's like those spin out things are like, what are they spinning out of? We've got it. No, I'm kidding. I'm joking. There's a ton of different segments. That was kind of our expectation about what was going to happen here is when we saw how explosive the growth was in the first three, four months, we were like, this is going to be everywhere and there's going to be a lot of different flavors of this.

• 3:51Auren Hoffman?

  Not to use the pun, but it was a bolt. It was literally like an explosive thing. I had never really seen anything like that before. The amount of the adoption, I don't think you even were doing marketing. Somehow people were like finding out about it and just signing up and starting to use it through kind of word of mouth or other types of channels. I imagine it was all hands on deck just to like get compute and tokens and things like that.

• 4:15Speaker 01

  Absolutely. We spent 90 days building the product. We had no idea it was going to explode like it did. So we were very unprepared for the giant surge of demand. Our team was 12 people at that time. In the first month, we went from zero to four million of ARR in the thing and tens of thousands of paying customers, again with a team of 12 people. And at that time, Anthropic was starting to scale up their first party inference. We were just slamming into the ceilings of the capacity that they had during that time. And then the second month went from four million to 20 million, which was even more nuts in the second month. It was crazy. That period of time was absolutely insane because we were so understaffed.

• 4:54Auren Hoffman?

  For the audience, this is literally a year ago today. This is kind of insane. Seems like this happened three or four years ago, but it actually was just a year.

• 5:03Speaker 01

  Just a year ago. And since then, it's still been just kind of like holding on to the tail of the dragon. We've continued to scale up, going and meeting demands is one thing, but then, once you kind of get to where the demand level was before, there's even more to go now because the demand's continued to grow. I've never seen anything like it in my career. I've been doing this for 15 years. It's been intense the past year to keep up with the demand and scaling with a small team. I think we've done a very good job on our team of doing that, especially in the first three months. It's wild that we kept everything online. We had basically zero downtime for the most part in the first 90 days, which is, considering the size of the team then, the scale that we were operating at is kind of unheard of. It's been crazy.

• 5:44Auren Hoffman?

  One of the other things I think is also completely crazy is you guys were the first real to explode, but the whole space is exploding. The whole space was zero ARR 13 months ago. And now I don't even know what the ARR of the whole space is, but it's in the high hundreds of millions of dollars. And then there's all these companies that are doing it, Stackbots and Bolt New being like one great company, but there are other great companies. And it seems like you're all, every week, one-upping each other. It's so competitive that everyone's making everyone better. You can't stop innovating because otherwise your competitor is going to eat your lunch. People are just going to switch to the other. So I imagine it's crazy. It's like playing in the NBA. You just have to keep upping your game.

• 6:27Speaker 01

  A lot of this year has been an all out mad dash. How fast can you ship and how fast can you innovate? And you're on the ground floor of like a gold rush. There's this frenzy that happens where some amount of it is warranted where it's like, hey, we need to go and secure strategic assets in reserves or whatever. But then there's a part of the game where things change and you start looking at things more critically. And I think we've begun entering that phase. Maybe a good example of this would be if you look at the United States of America as a landmass, when we first got here, this expansionism coast to coast, and it was just about purely acquiring land, it's pure land grab. But over the decades and centuries after, some of the most valuable land became only

• 7:09Speaker 01

  apparent as you dug in and said, what kind of businesses are going to be built on these different plots of land? And for example, in the case of Exxon, they don't really care about buying all of the land in America. They care about very specific plots of lands that they know oil is underneath that they can pull out and sell. I think we're kind of at that phase of the game where it's no longer about grab land at all costs. It's really more of who are the customers that you really want to focus on? There's tons of these things, there's tons of competitors, there's tons of things you can focus on. You really have to get specific on where's the oil? Where's your business? Where can you really have your unique slant? That is clearly where the game is going. I wouldn't be surprised if the other guys are saying the same thing.

• 7:47Auren Hoffman?

  How do you think about Anthropic? Almost all these companies are built on Anthropic and wouldn't really exist without the hard work that they did. Obviously, there's OpenAI, which is trying to build some things also to compete with it. How do you think of those guys?

• 8:01Speaker 01

  Anthropic's done an incredible job of really pushing the frontier of AI code job. They started to dabble with their own offerings. When I say dabble, it's I mean, Claude Code went from zero to $500 million of ARR in a couple of months or something. It's wild. That was not a giant, big product release. The surface area of that is super small. And it had a pretty material impact, especially on the more developer oriented tools. From what I've heard through the grapevine where when Claude Code came out, that changed the dynamics for growth and et cetera, some of these other things.

• 8:35Auren Hoffman?

  A lot of cursor people moved over cursor was taking from Git.

• 8:40Speaker 01

  Exactly. I look at Anthropic and OpenAI and whatever have you, and all of these companies are going to be getting more into the application layer. Outsider looking in, I think they probably want to own more of your direct customer relationships versus just pure API usage. They probably have some specific angles where they think they can add additional value in end user products through vertical integration that just selling their stuff via API would not be able to do. I don't think there's a reason they shouldn't do that. And I think as a startup like us or like Reflet or like, well, you have to look at that and go, what do we think is likely that these guys are going to do? What are they going to care about? And even some of the other incumbents, you look at Wix picking up Base 44, very small Bolt clone that they picked up. And they're doing a couple million of error when they got it.

• 9:26Speaker 01

  And they've pummeled their distribution behind it. And I've heard the numbers are extraordinary. I don't know if they're true, so I won't repeat them. They're up there with us and the other guys as far as revenue.

• 9:34Auren Hoffman?

  Cause they had a channel, so they basically bought something and stuffed the channel essentially. I imagine a lot of quote unquote website makers, they've never even heard of lovable or bolt new or reflet or versal. They're just like, I'm in Squarespace or I'm in Wix and I'm just making my website.

• 9:50Speaker 01

  I think that was a real question for us. When Bolt exploded, we started seeing the usage happening. One of the important signals that we were kind of looking for is are the existing web 2.0 site builder hosters, are they asleep at the wheel? Are they not going to do something about this? Because if not, that in itself, you could build a company that takes out Wix or Squarespace or something. Given how fast Wix moved and the prowess of the result there, I think in the consumer space, the existing incumbents are not asleep at the wheel. It seemed to be doing pretty good work. That's the job of when you're in one of these gold rush markets is you have to like delineate. There's opportunities everywhere, quote unquote, but a lot of them, you have to be ruthlessly pragmatic and go, do we really think that?

• 10:33Speaker 01

  This incumbent or this other company or whatever, you think we can really smoke them there? The cost of not doing that is very high. The cost of not thinking critically about that is extremely high.

• 10:45Auren Hoffman?

  It also really just figure out who is my customer? Who's my ICP? In some ways, you don't want to compete against six amazing companies. You want to really compete against one company. Let's narrow down and let's just figure out like, we're going to just be the best at this thing. And maybe we believe this thing is bigger than the other people believe. So they're not going to focus on it as much. I assume that it's going to eventually happen. It's just too hard to have six amazing competitors all beating each other with clubs.

• 11:13Speaker 01

  A hundred percent. This just strikes me as not possible. That's what we've been up to really lasering in and going, where are the things where we think that it's off the direct warpath of the other incumbents? It fits in with this strategic edge that we have as a company, or strategic edges that we have as a company, et cetera. How we really orient towards that versus a lot of the past quarter or two has just been pure top line error, growth, press, just attention warfare, da, da, da. Which we've engaged with and done, but when you kind of look beyond that, what's the real sauce here? That's allowing us to win. For us, it's a focus on selling to businesses and product teams. For us, it's been huge. We are the number one tool used by professionals at 72% of the Fortune 500. None of the other guys are anywhere close to that number.

• 11:58Auren Hoffman?

  And that might mean, okay, we have to build more collaborative features amongst people. It's not just for an individual, it's for a team. And then there's all these weird team features that one has to build in. And then, okay, what do you let people see? And what do you not let people see? And maybe there's got to be security around that. That's a whole other thing that maybe somebody like at Wix, they wouldn't build that because it's, I'm just focused on like the small business owner or something.

• 12:21Speaker 01

  That's been our angle. I think other people are looking in this direction as well. Again, there's so many opportunities. You don't want to say no to all these things, but in order to really go win, you have to, regardless of how much money you raise.

• 12:32Auren Hoffman?

  You have to figure out, okay, this is where we have clear right to win. We're going to build the best, best, best product for this group. We just got to also understand this group is a big market. This can get us to the billion plus in revenue and then maybe we're going to move to the next group. I love talking about a little bit about marketing. It's a weird thing. Usually when you have a company that's growing this fast, all these companies growing this fast, the company name and the name of the product is the same. But you have Anthropic and then they've got Clock. You've got OpenAI and they've got ChatGPT. You have Stackblitz and you've got Bolt new. It's very confusing. Now I have to remember all these other names. Why did it end up that way?

• 13:12Speaker 01

  Anthropic and OpenAI probably have a better case for it. Bolt was the last experiment we ran before we were going to start shutting down the company.

• 13:20Auren Hoffman?

  You guys were running out of money. I remember this. This is like five pivots into the company.

• 13:25Speaker 01

  Exactly.

• 13:26Auren Hoffman?

  I don't even remember what we funded, what we originally funded you back in the day. I don't even remember the original idea because I remember you guys were like these super smart guys. You were just looking for product market fit.

• 13:36Speaker 01

  Totally. So I mean, we built some amazing technology that was ended up being really the core of how Bolt works with our in browser dev environment stuff. But when we were going to launch Bolt, like we had a product called Stackblitz. I was like, the name of the company and the product were the same for Stackblitz, whereas a web based IDE. And when it came time to try out a new product type that was very different, it was the same core technology, but a very different product surface. I mean, experienced it before. We don't know if just adding a page on our site that will let people try it will be enough. We kind of think it needs to be its own standalone thing. We came with the name Bolt and we didn't think much was going to happen. Bolt's actually like what most people know us for.

• 14:13Auren Hoffman?

  Probably most people have that haven't even heard of Stackblitz.

• 14:16Speaker 01

  Exactly.

• 14:16Auren Hoffman?

  Maybe just change the company name to Bolt.

• 14:18Speaker 01

  We're probably going to rename the company would be my guess. We're probably going to align the branding. It's interesting. Anthropic has started rolling over even their API console is underneath the Claude.com domain now.

• 14:30Auren Hoffman?

  So Claude is really the big brand name there.

• 14:33Speaker 01

  They're really leaning in there on that. Then that's like recent. I guess it's actually not necessarily unique to us. As these different companies are finding product market fit with specific products and the names of them, they're going and they're saying, okay, well, this is what we're known for.

• 14:46Auren Hoffman?

  Cause a lot of people I know who use Claude don't even know the company name is Anthropic, but they don't even know what it is. So like, I'm just using Claude.

• 14:54Speaker 01

  It makes sense. It seems like that's maybe a common thing. I think for OpenAI, it's like ChatGPT. I guess GPT is in the name. That's a hard one though.

• 15:01Auren Hoffman?

  ChatGPT is such a terrible name. No one knows what GPT stands for. It just seems weird. It's very, very confusing.

• 15:10Speaker 01

  And it doesn't matter. That's what's incredible. That's why I love it.

• 15:13Auren Hoffman?

  Exactly. That's the thing. It doesn't matter. Usually the name actually matters a lot. We have this whole theory at our firm that we just look at the names of the companies when they're raising. And if it's a bad name, they're never gonna make it. We just have a whole history of smart people with bad names that have never made it. The name actually matters quite a lot. But somehow, if your product is so good, it really doesn't matter.

• 15:37Speaker 01

  Totally. That was an interesting thing too with Bulwark. We had to cut so many corners when we were building. We had 90 days to build the thing. We didn't have a mobile responsive view when we launched. And we didn't have a mobile responsive view really for like a quarter after we had launched, but we got to 20 million of ARR with no mobile. There was no mobile version of this product.

• 15:54Auren Hoffman?

  That makes sense, given the personality you're going after. They're on a laptop. That's the key.

• 16:00Speaker 01

  It does, but we were getting cooked by people on Twitter because they were taking screenshots from their phones like, dude, what is this? How do you guys not have this? Well, it kind of goes to show like the conventional wisdom.

• 16:09Auren Hoffman?

  If you're building something, you're building something on your machine, not on your phone. Builders don't build on phones. They build on their machine.

• 16:17Speaker 01

  We had to do it because we didn't have the time, but you go to any startup and you're like, we're going to launch this product. It's going to be people to make software on the web, blah, blah, blah. There would be good arguments of, listen, if you don't have a mobile responsive view, no one's going to put in their credit. This is so unprofessional, like blah, blah, blah, blah, blah. The brutal pragmatism, and that's to me a good example of the naming thing too. Over the past 15 years of being in the valley, I think there's a lot of this conventional wisdom that's, if you're brutally pragmatic, it's meant to be broken that because you save yourself a ton of time that would have been wasted or would have delayed our launch and, which maybe changed our destiny if we wasted time on polishing those corners.

• 16:53Auren Hoffman?

  Yeah, you might not be around.

• 16:54Speaker 01

  Exactly. We may not have been one of the first to market there. And I think other companies might have taken them too. And there have been companies that have launched. They came out six months after us and Base44 was one of these that got, and then they did great acquisition for that founder, by the way, $80 million for a single person working on something for six months.

• 17:11Auren Hoffman?

  By the way, it sounds basic.

• 17:12Speaker 01

  That sounds awesome. You did a great job, but just to kind of go and show though, at time of acquisition, our revenue was 10x or something. It was flatlined. Speed matters and rules are made to be broken, but you have to be brutally pragmatic to do that.

• 17:26Auren Hoffman?

  World of DAS isn't just a podcast. World of DAS membership is a private invite-only community for founders and executives of important data and AI businesses. It's a place for off-the-record, high-trust conversations focused on the biggest questions data leaders are facing. The community is about peer learning, curated intros, and access to private events. Everyone in the group is building or leading something meaningful in data. If that sounds like you, apply to join at worldofdas.com. That's world of D A A S dot com.

• 18:07Auren Hoffman?

  When you first started, it was nine bucks a month, some crazy low number. And then I assume you were quickly were like, oh, my margins are negative because you're spending money on tokens and stuff. And then you have to like, figure it out. How'd you figure out the pricing so quickly too?

• 18:23Speaker 01

  There's a story here where I almost had a heart attack or something. We had 90 days to figure this out and we had not had a lot of experience. We were just trying to figure out, is there something interesting here enough to keep this company alive and worthwhile? We're like, hey, if we had a hundred thousand dollars of error by the end of the year, that would be crazy. Best case scenario was that. So going into it, we were like, on Stackless today, we have one subscription plan. It's nine bucks a month. So for the sake of time, let's piggyback on that. Let's launch this and if there's feedback on it, we'll, we'll circle back. Nine bucks doesn't get you a lot of inference.

• 18:55Auren Hoffman?

  No. My chat GPT subscription is 200 bucks about personally.

• 19:01Speaker 01

  Totally.

• 19:02Auren Hoffman?

  It's still slow.

• 19:03Speaker 01

  When we went to launch it, we were like, well, we want to make sure people get enough. Let's give them like a little bit more than nine bucks of inference to use, actually. I can't remember how much it was. We were going to be bleeding money on people that came in. We were like, well, they probably won't use all of their usage and we want to prove that there's demand and we can like adjust pricing or whatever. We launch, it explodes. We have tens of thousands of people signing up, burning through inference. They burn through their nine dollars, a lot of these people on the first day or two. And they're like, I want more and we have no way to give them more because we have one subscription plan. People are screaming, I want to use your product more. I want to use your product more. I can't use more. People are signing up from multiple accounts just so they can keep going and using things. I sat down with the team and I was like, I will work through this the entire weekend with whoever is around to get

• 19:48Speaker 01

  higher tier subscriptions landed. We have to do this. We worked literally around the clock, 24 hours a day, got it done.

• 19:57Auren Hoffman?

  The best scenario, which rarely happens is people being like, I want to give you more money. And then you're like, I don't know how to take it.

• 20:04Speaker 01

  Exactly. I rarely ever ask our team to be like, hey, we need to work around the clock. You don't want to burn out your team, but in this scenario, this is a moment that we have to seize and I will do whatever it takes to get it done. And we did. And the other thing too, is at that time, usage based pricing is now a very common thing. One year later, we were the first company to do it.

• 20:24Auren Hoffman?

  You were the first? Oh, wow.

• 20:26Speaker 01

  We were the first company for AI code gen. Cause everyone else before this was 20 bucks a month, all you can eat. But if you eat too much, they were throttle you, blah, blah, blah. You couldn't pay more to just get more. We were the first company that said, screw it. If you want more, you can pay more. You can just buy more. Everyone does that now. When we did that, I think the month after we rolled out that pricing, Barclays actually did a report. They were like, this is the first time we've seen this from the outside in. It looks like it's working insanely well. We think this is going to become the new default and that's exactly what's happened over the past year.

• 20:55Auren Hoffman?

  I imagine all of a sudden you have these two customers that are spending just gobs of money. You're like, how are they spending so much money on us? Did you start calling these guys up? What are you doing?

• 21:06Speaker 01

  Absolutely. Cause we were like, what is going on? Just the explosion of people coming, who are these people? Cause we thought developers were going to be the ones using this product. That we've been building developer tools and we thought it was developers were going to be the main audience. And when we went and talked to these people, they weren't developers. They were PMS. They were product people. They were startup founders. They were sales guys. They were every walk of life. It was kind of like a murder mystery. We were like, okay, what the hell? What is causing all of this? It started to click when we talked to these people like, oh, okay, wow. This is allowing non-developers to actually build and ship software. Okay, that changes everything. The TAM here is not 25 million developers. The TAM here is a billion people, which was mind blowing.

• 21:43Auren Hoffman?

  This is crazy. It's insane. At the same time, you're building these tools for people. You're also using all the code tools to build these tools. I assume you're using all this other stuff as well. What have you learned as you're building an AI coding company for other people? But you're also building internally an AI coding company for yourself.

• 22:04Speaker 01

  Totally. It's all about how do we maximally leverage the resources we have for the maximum amount of scale that we can achieve. Certainly, we're using code gen tools ourselves. Like we use Bolt internally, we use Cursor and Cloud Code and whatever internally. The other really key unlock was during that first three months when we were exploding in usage with a 12 person team, none of which was customer support. My chief of staff and I were responding to literally a thousand emails a day for the first two months. No one else. And I called up my buddy who is an MD in between jobs or whatever. And I was like, we need help. I will pay you money to come help write these support emails.

• 22:39Auren Hoffman?

  He's a doctor?

• 22:40Speaker 01

  He's a medical doctor.

• 22:42Auren Hoffman?

  That's hilarious.

• 22:43Speaker 01

  And now he's working full time with us. Actually, he's doing other stuff with us, which is hilarious. We call it, I have my wife responding emails, et cetera. But then with this company, ParaHelp, there were two guys back then. And they probably had a few more people at this point. But why company or company? They're doing AI based support that pulls your documentation from all of your Discord, whatever, to answer questions and support and refunds or whatever. We integrated that thing and that thing takes out 90% of our tickets automatically. It's the top rated support agent that we have. Our support team today is just under 10 people. I think it would have to be like 50 to 100 people just doing support.

• 23:20Auren Hoffman?

  You'd have this whole office in the Philippines managing all these people and stuff.

• 23:25Speaker 01

  And the quality support would be lower. This is an AI agent that's powered by GPT five or something. If you ask it a technical question, I'm getting this error. It actually will be able to like, kind of search the web and be like, oh, this is maybe what's going on. It can actually do the job of someone maybe at almost the level of a sales engineer almost. We're constantly looking for things where it's like, are there tools that we can adopt that allow us to scale up without adding a ton of headcount and allow it to be much higher leverage. We're constantly looking for new AI tools or startups or whatever that we can bring in.

• 23:57Auren Hoffman?

  If you hire too fast, it can be really the death of a company like yours. Managing all these people's heart, you have this N squared communication problem in any company. You don't want to hire too many people too quickly. So you've got to bring in all these other tools to help you get everybody leverage.

• 24:12Speaker 01

  Absolutely. Having a small, tight knit team allows you to just move faster than going and scaling up headcount. We lived through the 2021 era as a company and we did not take the bait of just raising a boatload of capital and just deploying it and hiring people. And the founder status games are like, what's your office? How big is it? How many people do you have?

• 24:33Auren Hoffman?

  You're so lucky you didn't because you would have been out of money. You would have never made the pivot.

• 24:37Speaker 01

  Exactly.

• 24:38Auren Hoffman?

  We flex capital. We invested in you guys in 2019, which is

• 24:44Speaker 01

  The original seed.

• 24:45Auren Hoffman?

  That's six years ago. It's so long ago. You're kind of in the wilderness for a while pivoting. Now you're a huge mass of success. So now it's awesome. So now we can kind of laugh about it and talk about it. How did you keep the team motivated for those five years? You had these amazing people who easily could have gone somewhere else, made a lot more money. Somehow they stuck with you, maybe not everybody, but a very core group stuck with you that whole time. What did you do?

• 25:11Speaker 01

  We were really a group of engineers. I just wanted to build some really cool technology and bring it to market. Stackblitz was basically a research lab in the first four years of the company that was trying to figure out how to bring software development environments into the browser using WebAssembly. That was what we were doing. And at the same time, we were releasing our products, like our web-based IDE, going competing in that space, et cetera. In 2021, we finally cracked it, the technology. We put out the first version that was really usable. It was a big deal in the software development world. The real problem though is two years or whatever after we launched it, we couldn't figure out how to actually make money on the thing. None of us were working on this thing because we actually expected to make

• 25:55Speaker 01

  A ton of money, which sounds weird, but the reason we're here is just to build cool technology. That intrinsic motivation is what allowed us to build it and persevere. And that technology we made, Web Container, is what powers both. What allows us to have healthy, sustainable margins where our competitors are having to bleed money on cloud VMs and stuff. That's a huge advantage that they will take years if anyone else wants to go and build something similar. To me, it's like if I were to start another company, I think it's always good to just get mission-driven people. If you're working on something because you really just love it and you're making progress, which we were, you're not going to want to go. If you're working with really cool people and a really cool mission and you're making progress, it's not just superficial. Oh, man, we're in the front page of TechCrunch

• 26:39Speaker 01

  today. The past month was crazy. We had no idea how to make a file system work in a browser. Now we do. These tangible milestones that are cool to you, that maybe to others that are playing different games, they're like, what? That's kind of the nature of long-term games. Look at professional athletes. I grew up in Chicago during the 90s. Michael Jordan was just at practice, man. That was his thing. He was going and doing extremely unglamorous physical work most of his waking day. If you want to be really great, you have to be willing to do that for extraordinary amounts of time that most normal people wouldn't be willing to. Everyone wants to be an NBA star. Not many are willing to do actually what it takes to do that.

• 27:17Auren Hoffman?

  It's freaking grinding.

• 27:19Speaker 01

  Yeah, and bucking the social games that come with these things.

• 27:21Auren Hoffman?

  Especially you live in San Francisco, you took some venture money. It's hard to not play the social games. You get it all around you. Every coffee shop, every social gathering, every time you look at LinkedIn or X or something like that. And obviously you can't avoid it 100%, but how do you not get sucked in?

• 27:41Speaker 01

  I don't go to a ton of networking. I'm actually kind of a recluse. That's the funny thing. I live in the Bay Area. I don't go out a ton. I've got a wife, a daughter, and a dog at this point. I like spending as much time as I can with those folks. Even before this, we've been a remote company, so we haven't had a giant SF office. We were about to open one up, but even then, we're not planning to do to ramp a ton of hiring. We're really going to be using this office for a lot of marketing and things of that nature.

• 28:08Auren Hoffman?

  Sales meetings.

• 28:10Speaker 01

  Totally. And maybe it's just the nature of how I approach solving problems. I like sitting and thinking on problems for long periods of time. I guess I'm the sort of person, when I'm analyzing stuff, I don't really go and talk to Thirty people. I'd like to come to my own conclusions and it's actually kind of important to me, I guess. Not that I don't ask for help or ask for questions and that sort of thing when needed, but if you could just kind of get the answers from conventional wisdom and just from the crowds around you, that would mean you have effectively almost no edge in finding insights that are valuable. And that's just not the case. I think you have to do independent thinking and make the space for that. And I think I may be a little wired towards that. I would avoid networking events. I'd spend more time just talking to your customers and thinking with your team, thinking with yourself.

• 28:52Auren Hoffman?

  I think avoiding networking events is always a good suggestion, unless you're like a sales guy or something. So speaking of sales, I mean, you basically have at least four things that you've got to be focused on. You're focused constantly on product, making the product better, at least for that core group that you care about. You've got to somehow be focusing on your margins. This is a very tricky world where they can get out of hand and you have to make sure you've got a robust business. You can invest back into the business. You've got to be focused on marketing. So you can put all these new people in the funnel. And then maybe you got to focus on enterprise sales, which is probably a very new thing. And that might be a weird product that we have to do, or there might be a service contract we have to do on top of that.

• 29:34Auren Hoffman?

  How has all these evolvements happened over the last six months?

• 29:38Speaker 01

  It's been a lot of stuff running in parallel. So we brought on our head of sales at the beginning of the year. We were going to B two B side before anyone else in the space was really meaningfully looking there. We've been investing in that end a lot. A lot of our work on the product and engineering side. This phase is moving so fast. That has been the primary place that I and a lot of the team have had to really, really laser it on and just scale the size of the team and go and ramp our current technology. And we just put out our V two release almost exactly a year after we had launched a couple of weeks ago. We've got the state of the art of anything out there now. The fruits of that labor have very much panned out for us and there's a lot more work to do on those things.

• 30:20Speaker 01

  We tried a lot of different experiments around consumer marketing versus B2B and all of that culminated over the past quarter, really in our decision to really laser in on B2B use cases. We have our sales, our sales engineers, folks, our EPD teams, they're spending a lot of time with customers. How are they using the product? How do we really lift the ceiling of the types of things they can do with this for their ambitions? It's been very collaborative and every week is different challenge and different insights.

• 30:48Auren Hoffman?

  And how do you think of capital as a lever? Earlier this year, you raised like a hundred million dollar Series B. How do you think of that as a lever? Obviously, lots of other companies in the space also have a ton of capital as well.

• 31:01Speaker 01

  To a certain degree, you need some amount of money to be playing at this level of the game. I think for the scale we're at, that's why we raised. If we're going to continue to go and meet this scale, we have to have this type of capital on hand.

• 31:12Auren Hoffman?

  You gotta sign commits sometimes. Those commits are big and people have to understand that those commits are serious.

• 31:20Speaker 01

  Absolutely. When we did our first commit, it was the scariest thing I'd ever signed. This was October of last year. I think we committed to like three or four million dollars of spend or something for an entire year.

• 31:31Auren Hoffman?

  And at the time you have four million ARR or something.

• 31:34Speaker 01

  Of ARR. I think we ended up going down to like two or three, something like that. But it was still like, Jesus, this is a ton of money. This might be me signing the grave of this.

• 31:42Auren Hoffman?

  Then you're like, oh, we spent it in a month. We thought it was going to take a year. We literally spent it three weeks later.

• 31:48Speaker 01

  It's one of those things where now the commits are even bigger and it's important thing. I think the other thing too from 2021 where when we looked around at the companies that were raising just tons of money, it kind of breeds this bad mindset where it doesn't mean that companies shouldn't raise a lot of money if they know how to deploy it effectively and blah, blah, blah. But the problem is that if you are running into problems with deploying it effectively. Then you start just doing things that sound good that will maybe get you likes on LinkedIn. Like we just hired this heavy duty person that's now going to hire 50 more people underneath them. You start doing these things, not say that that's not what a company should do, but you're doing things that are further ahead of where you are as a business, remove how much money is in the bank account. You, product, business, customers, where are you at?

• 32:33Speaker 01

  But once you raise a ton of money.

• 32:35Auren Hoffman?

  They're like, oh, you need like a world-class CFO. I'm like, why do I need a world-class CFO?

• 32:40Speaker 01

  Exactly. And the stuff slows you down because the CFO's job is to make sure that nobody's going to go to jail, which is an important job, especially if you're going to prepare to go public and blah, blah, blah. If you're not there yet, it's going to slow you down from getting to where you actually would need that guy.

• 32:54Auren Hoffman?

  I love the anecdote you had about the customer support. Because you're right, you could have just spun up all these people. And then of course you would need to hire these executives in customer support who know how to spin up all these people and motivate these teams, or you could just work with another company that's already very good at that. Maybe they're not exactly what you want, but they do a pretty good job. They can take off 80, 90% of what you need. Maybe you're even paying them more. It's not about the money. You got rid of it. There's no time. You don't have to spend all this time hiring people. You don't have all the complexity of all these extra people in your org that you have to motivate and give them a career path and all those other types of things. And I think more and more companies like Stackblitz are doing this, which is really exciting.

• 33:39Speaker 01

  We didn't raise it too much where we would be going. Okay, we're getting external pressures to do things because expectations are set so much higher. But relative to where the business is at, people don't want to wake up to the reality of that. We may start to see some of that happen here in the AI space where companies have raised a lot of money. But they're finding out, okay, wait a second, this business actually isn't as mature as we thought it was when we invested. And that's where these bad paths can kind of happen. A lot of the 2021 zombie companies sprouted out of. We've kind of done a good job over this past year, but also in the previous years of calling the balls that we actually want to swing at and connecting in a way that's going to allow us to sustain and grow.

• 34:19Auren Hoffman?

  A few personal questions. How are you personally using AI in your personal life, with your family or with yourself or hobby? How are you personally using AI?

• 34:29Speaker 01

  I use ChatGPT a lot, Google's AI mode stuff, whether it's random stuff around the house where I'm like, hey, this thing is, I think it's broken. How do I fix this faucet?

• 34:39Auren Hoffman?

  Oh my God, so good. You see this weird error on your dishwasher. You're like, I have no idea. Take a picture and oh, just do this weird keystroke thing. Wow, this is amazing.

• 34:50Speaker 01

  Yeah, it's like Dad GPT. Call my dad and be like, hey dad, have you ever like fixed one of these things? Sometimes I still do if ChatGPT doesn't know how to do it or whatever. But stuff like that, a lot of life is about figuring out how to calibrate your perspective. My experience as an entrepreneur is an example of this, but if you're like trying to go and find, hey, we need to hire a head of sales. There's tons of information. You can go ask people. They're going to have tons of different advice because it's so context specific to your company, your exact circumstances. Much of this advice is going to be useful, but it's going to be slivers of different things. It's going to be then this package of context for you specifically. This is like the sort of head of sales that you need at your company

• 35:32Speaker 01

  for your stage, for the type of product you're making. And you want them with this title, with this level of experience. Some of this is applicable, but a lot of it, you just need to like immerse yourself. Talking to ChatGPT or like whatever about this stuff is incredibly helpful because normally you would have to spend a ton of time talking to other entrepreneurs, sales leaders, which I think is still a good thing to do. But being able to just go and get a perspective to calibrate and ask the questions that are ancillary, but help give you a sense of the shape of the problem and the nature of it and the nature of the solution, the shape of the solution that you probably want for this and other things in life. I think it's great.

• 36:07Auren Hoffman?

  I find it great for like self-directed learning. I'm trying to learn about tariffs. I'm super interested in learning about tariffs, how they work, what's going on. And then I can start asking questions. Sometimes it tells me things like I clearly already know I can just cut it off. But sometimes it's going in a direction that I don't understand. Like, I need you to explain this more like a fifth grader here to me. I'm not getting what that is. So it's a very helpful tutor on stuff.

• 36:31Speaker 01

  That is what I use it for all of the time. And then like what I'm describing to you there is it can tutor you on things that are so context specific. You can just go back and forth, give it all the context. That's just never existed before. That would require you talking to other humans, which takes a lot of time to go and find the right people, get interested. Hey, do you have 30 minutes next week? That is such a process. Whereas if you can kind of get 80 20, you need to know that all the sauce, but you at least can be 80 20 on the synthesized result. And then you talk to the experts and they validate and then add their chairs on top. It just pulls forward stuff that would take so long.

• 37:04Auren Hoffman?

  When you talk to a human, there's some pleasantries you have to have. There's a two way street. You can't just take from them. They might be asking you questions. It's not as efficient. I can just be cutting your chat to you. I mean, hopefully the AI is still likes me when they run the world. I can be like cutting them off all the time and start moving.

• 37:24Speaker 01

  That is the primary thing that I use it for all the time because I have curiosity about the world around me and the things that I am responsible for in my personal and professional life. The more that I can be aware of that stuff, the better. I mean, as living creatures, we kind of tend to do in play what actually is important for us to survive and not play. Asking questions and understanding your environment is play, but it's important for where all of your personal alpha is and whatever endeavors you want to do.

• 37:52Auren Hoffman?

  Now, two more questions we ask all of our guests. First is, what is a conspiracy theory that you believe?

• 37:57Speaker 01

  I think it's weird what happened with Epstein.

• 37:59Auren Hoffman?

  Super weird.

• 38:00Speaker 01

  I think all of America is aligned on that. I don't know if that's even a conspiracy.

• 38:04Auren Hoffman?

  I still think there's a lot of good theories about what was going on, and I don't think there's alignment of what those theories are. There's probably like 10% of America that believe each of the weird theories, some of which may be true.

• 38:16Speaker 01

  It seems improbable that this guy died on his own.

• 38:18Auren Hoffman?

  I was like, how did he make his money? No one knows. It's just crazy. Clearly wealthy guy. No one has a sense of how did he make his money? Why did people pay him for his services? It's not even clear what he did. It's so weird. I'm with you, but I don't want to talk too much about it. I want my family to be safe.

• 38:37Speaker 01

  To be fair, I haven't seen the evidence. I could have my mind changed.

• 38:41Auren Hoffman?

  Exactly. All right. Last question we ask all of our guests. What conventional wisdom or advice do you think is generally bad advice?

• 38:48Speaker 01

  If you're building startups and you're in the valley, the hype sphere of stuff, anytime there's some narrative going on, I usually will tend to look at it and I'll really analyze why it would be wrong and really try and figure out what am I missing here? And oftentimes, there's kind of these gaping holes. People will follow these trends just blindly.

• 39:06Auren Hoffman?

  Well, we're mimetic creatures, so we like to copy each other.

• 39:10Speaker 01

  A hundred percent. Getting out of the rat race entirely and just spending time independently thinking about stuff and coming to your own conclusions is key. That's the thing about advice and whatever is that it's not all good. It's not all bad. It's often context specific. Yeah, I can help with delineating that stuff. I don't have anything specific because it all comes down to context typically. That would be my guidance on how to delineate it.

• 39:32Auren Hoffman?

  I love it. It's great guide. It's great advice. You're an amazing entrepreneur who I admire. Thank you, Eric Simons, for joining us on World of DAS. I follow you at Eric Simons on X. I definitely encourage our listeners to engage you there. This has been awesome and a ton of fun.

• 39:48Speaker 01

  Awesome. Thanks, Arun. Appreciate it.

• 39:53Auren Hoffman?

  Thanks for listening. If you haven't already done so, please subscribe to World of DAS wherever you consume your podcasts, YouTube, Apple, Spotify, and more. And please help us get discovered by leaving a review and check out worldofdas.com. That's worldofdas, D-A-A-S, dot com. And of course, connect with me on Twitter, at Auren. That's A-U-R-E-N. Would absolutely love to hear from you.