S13E2: Scaling Warp and Supercharging Developers with AI with Founder Zach Lloyd

don't grow before you have retention. It's very hard to grow something if you don't have a product that's delivering value and that people want to stick around with. Before we launched anything to the public was like, can we dogfood this? Can we use it? Is it actually helpful for us? This was, I'm talking like 2021. And so we did a, a landing page, just like a pure waitlist landing page, and we posted that onto Hacker News.

Welcome back to Season 13 of The Room Podcast. Claudia, can you believe we've been doing this this long?

No, it's been 5 years. So many episodes, so many incredible conversations telling the stories of some of the world's most iconic founders and funders. And whether our listeners are new for the first time today or have been with us from season 1, we're so grateful for the community we've built.

Completely. We've had the mission to open the door to the room where it happens for those entire 5 years and have been able to do so over 120 episodes with hard-earned lessons from some of the top technology and consumer founders of our time.

And we've been through a lot personally as well.

We have indeed.

I built, scaled, sold my startup Priv. You were our first investor, so you've been there through the entire journey and you've also recently graduated from HBS. You're now building at Vercel, and through it all, we've brought the funder and founder perspective.

Yeah, it's really been an incredible journey, and I'm thrilled that I'm back full time in San Francisco and we're probably here to stay. That means that there's more for us in terms of growth, both inside and outside of the physical room. But what's important to have really out there is these stories, and we've been able to do so over incredible guests from the CEO and founders of Flexport, Perplexity, Particle, Zapier, Zillow, Vanta. I could go on. These guests and their enterprises have generated over $100 billion enterprise value and really, frankly, moved the markets.

And growing, especially in 2025 with everything that's going on in AI. So I'm so excited to share Season 13 stories that are maybe a little bit more AI-focused than past seasons.

Absolutely. And you can find us IRL and URL, especially with our events such as our annual conference, Inside Summit, or upcoming SF Tech Week events and beyond.

And if you want to get info information around those events, subscribe to The Room Podcast and subscribe to our newsletter at theroompodcast.com.

Perkins Coie supports the most innovative entrepreneurs and investors in fast-moving and high-growth sectors, addressing their myriad of legal needs. But the firm doesn't just provide end-to-end legal and business counseling to its startup clients. It also facilitates introductions to key advisors and sources of capital. Perkins Coie's interactive website, Startup Percolator offers access to programs, resources, and rich dynamic content designed to assist entrepreneurs on their startup journey. To learn more, go to startuppercolator.com and perkinscoie.com.

This podcast is brought to you by Mercury, the banking platform businesses like The Room Podcast use to simplify their finances. It's time banking did more than hold your money. Now it can. With Mercury, you can pay bills in seconds, close the books faster, and even send invoices. Not only does Mercury do away with a patchwork of tools, it eliminates guesswork, giving you complete and accurate visibility into your business's finances, all from one account. Apply in minutes at mercury.com. Time is every developer's most precious resource, and on today's episode of The Room Podcast, our guest Zach Lloyd is on a mission to give it back. Before founding Warp, an AI-powered terminal redefining how developers write and ship code, Zach Lloyd spent nearly a decade shaping the tools millions of us use every day. As the engineering lead for Google Sheets and the broader Google Docs suite, Zach learned firsthand what it takes to build products that change how people work. But his path to Silicon Valley wasn't linear. Zach's career had taken him from NASA systems analyst to philosophy student to Yale Law dropout before ultimately finding his calling in building tools for builders. With Warp, Zack is taking on one of the most fundamental and untouched parts of software development, the command line. His vision: to make the terminal collaborative, intelligent, and fast, supercharged by AI. In this conversation, we talk about the moment Zack realized the command line was ripe for reinvention, how Warp grew to hundreds of thousands of developers with over 20% weekly growth, his take on the future of AI developer OS, and what it takes to fundraise from investors like Sam Altman, Sequoia, Google Ventures, and Mark Benioff. Zach shares his honest lessons from building through pivots, scaling through hype, and staying focused on the long game. Let's open the door. Zach, thank you so much for joining us on The Room Podcast today.

I'm excited to be here. Thanks for having me.

We're excited to learn a bit more about your story and the story of Warp. But before we get into that, we really love to start at the beginning with our guests. Tell us, where did you grow up and how did that shape your view of the world?

I grew up in West LA. I kind of ended up thinking that there were a lot of celebrities around. I had a bunch of celebrities in my high school, which I guess is not proven to be true in my adult life. I don't know. I had a pretty normal, I think, upbringing and pretty fortunate upbringing. I would say like got good education in LA and I liked my childhood.

Not everyone has celebrities in high school. Who were they?

Jake Gyllenhaal was in my high school class. Jason Segel was a year above me. I went to this pretty fancy high school in LA called Harvard-Westlake, which a lot of people who are in the entertainment industry send their kids to. And so, you know, it was one of these unusual high school experiences where I think rather than like athletes being the stars, it was the people in the drama program who were like the cool kids.

That is a stacked yearbook for sure. But incredible to see that you are in fact a celebrity, but in the tech ecosystem.

I don't know about that.

In that vein of early identity, you have the celebrity ecosystem, folks going from their school play to Hollywood movies. Did you always see yourself becoming a founder or CEO someday, or was there a specific moment where that possibility clicked for you?

When I was in college, it was most likely that I was gonna end up doing something academic. I always thought I would actually go get a PhD. The area I was interested in was actually like philosophy and logic. Even though my undergrad was in something where computer science was a big part of it, it was this hybrid degree where it was computer science and math and logic and philosophy. I didn't actually think I would end up starting a business. And in general, I would say it took me probably more time than I feel like most young people take these days to figure out what even area I wanted to spend my time on. So, it took me a little while.

You're certainly, you know, a jack of all trades, have really explored a lot of pretty interesting sectors. Went to Stanford for undergrad, then you were a NASA systems analyst. From there, you got a philosophy degree and then went to Yale for law school.

Yeah, I was all over the place for sure.

Tell us a little bit more about how these early years really like shaped that career path for you. And also, I'd love to double-click into law school. Like, why did you leave after one year?

Yeah, my mom would like to know as well. When I was in college, I had broad interests. I liked doing technical stuff. I really liked, you know, I'm pretty like left brain oriented, I would say. So I really liked anything that was analytical. Actually, the thing that felt like the best match for me was logic, which is to this day, I find super interesting. How does logic map onto getting really clear about how you think things through? What can you derive if X But logic itself is a great foundational thing to study, but not really a career. There's probably 30 logicians in the world or more than that, but on like order of hundreds of people who do that. Because I had a wide range of interests, it really took me a while to figure out, did I want to do something that was technical? Did I want to do something that was just analytical? And law, I think a lot of people who really aren't totally clear on what they want to do end up in law school as a sort of default path. Or at least it used to be that way when I was coming out of college. Since I didn't have total clarity, law was pitched to me as like, hey, you know, you can do a lot of different things if you have a law degree. And so I ended up applying to law school and getting into a great law school. And I was like, I'll go try it. But when I was there, I ended up feeling like it wasn't really an area of super high passion for me. And even though actually I liked law school fine, I became skeptical that the sort of careers that people typically do after law school, which for people at Yale are either like, go work at a big law firm or you go work in policy or government or something like that. None of those things were as appealing to me as the notion of going and trying to like build something. So I ended up first putting it on pause and then finally just being like, hey, this isn't actually what I want to do. I'm going to do something else.

From your perspective, it probably felt like a winding path of what am I interested in? I think a lot of the founders we've had on the podcast cast have gone through that winding path, and it's really those like multidisciplinary experiences that really inform their unique alpha as a founder in many ways. So incredibly cool to hear that was your experience. So after law, you go into engineering, notably spending almost 8 years at Google as an engineer. You were the overall eng lead for Google Sheets, Google Docs suite, a pretty important set of products that you were building that pretty much everyone uses every single day. Tell us, how did your time at Google shape some of the insights behind Warp?

So my time at Google was awesome. I was really lucky to randomly end up on the Google Sheets team when it was about 5 people. It was what was called like a Google Labs product at the time, which Google used to have this kind of like area where they would release things. They weren't totally sure if they were going to be like part of the official Google Suite. They would try them there. And so I got into that. Project very early. It was a great group of people. So super smart people. My boss there, I was very fortunate with him. He's now the CTO at Notion and has done a lot of cool things in his career. And it just turned out that what we were building was something that was really valuable to people. And it was valuable primarily in like how it saved knowledge workers' time. We were working on a thing that everyone uses. So if you're going to be working in any kind of knowledge worker field, you're going to be using a spreadsheet. And it had a really killer feature, which was that it was available wherever you were by link. You could have multiple people in at once. You didn't have multiple versions of it floating around. And it was cool to be very early building something that was really useful for people. And so when it came to Warp, I actually was looking for something that had similar characteristics. And Warp is a, it's a developer product. It's not like a general knowledge worker product, but as a reimagination of the command line, the command line is similar in that it's also used by every developer every day. You know, I, I kind of felt from the Google Docs experience, if you built something that was significantly different, significantly better, really improved people's productivity, super fun experience going from something which hardly anyone used to something which is, you know, on Google Docs is now used by hundreds of millions or actually maybe even over a billion people, which is pretty cool.

That's incredible. And so cool to see how the structural philosophies behind what made Google Sheets and Google Docs successful be a thread that you pulled in your startup today before we spend a bit more time on Warp. Warp was not your first startup. You were the co-founder and CTO of venture-backed startup Selfmade. Tell us, what did Selfmade do and how did that experience influence your second startup?

Selfmade did a bunch of different things. The, like, original vision behind Selfmade was a photo editing marketplace. And so the idea was people care about the quality of photos that they share with their friends or if they post online and that people would pay to have a photo editor make their photo look cool or good or retouch it, something like that. That was the original vision. It turned out it was hard to make that like work. There is demand for that, but there's also like a lot of competing products. And so we're competing with apps that people can download and just edit their own photos. And the other thing is it was just hard to make the economics of that work well. It's expensive to pay a photo editor to make a photo look good. And so over the course of the time at Selfmade, we basically pivoted it to become much more of a marketing company where rather than trying to edit photos for individuals, small businesses would hire Selfmade to do their entire social presence, a digital marketing agency, but with a bunch of technology behind it. And again, there's a lot of demand for that. So if you are a small e-commerce brand, was like our best customer towards the end, and you want a presence on Instagram, you know, you're totally happy to pay $500 a month if you get a great thing that will drive business to you. I will say this was not the world's best fit for me. My interests are really going back to time at Google. I really liked building these like horizontal tools that are much more software oriented that solve problems that I personally have. And like, I don't even really use social media, so it was a great learning experience and like I got a lot of out of it. We managed to get decent revenue and growth, but it was hard to make it work at venture scale. It had a huge operational component. And so my big takeaways from that experience, I think really informed how I wanted to approach my next thing, which was to focus on problems that were a little bit better founder fit for me, I would say.

This is not the first time we've really heard this story of first startup and then realizing that this is something you're committing to for potentially decades. And so if you're not passionate about the problem or it's not necessarily the right founder market fit, it might be a better use of your time to keep building, but in a space that you wanna dedicate your life to. And so I'm hearing that first startup was an incredibly valuable time to learn that lesson.

That's right. I learned so many lessons from that startup when I ended up Moving on from it, I spent the first few months just like writing up a personal, call it a postmortem or learnings thing of all of the, just while it was fresh in my mind of all of the things that I got out of it, both good and bad that I could apply to my next thing. Yeah. And one of, one of the big ones was really to work on something that was a problem that I wanted to solve because it's, it's like you said, you're doing this for a really long time, even in the best case scenario, you're doing it for a long time. And then it's like startups are really hard. And the thing that kind of gets you through when it's hard is do you care about the problem that you're solving? Do you feel passionate about it? And so that's important.

I'd love to dig a little bit into some of those insights from that postmortem. You founded Warp in 2020 with the mission of really helping developers ship better software more quickly now through AI-powered command line experiences. Why was this problem space so exciting and important to you?

When I was going to found something, it was either going to be helping knowledge workers or was going to be helping developers. Those were the two areas I felt that I fit into. When I was thinking about what I wanted to spend my time on, I ended up thinking development was really interesting. A lot of the tools that I use as a developer are rough product-wise. They tend to be built by developers. They don't have the level of craft that you might get from, I don't know, using like Slack compared to like using ICQ or whatever. And so I thought there was a big potential to just try to build a great product company for developers was one of the insights. When I looked at the specific tools that developers use, there's two tools that developers spend their days in. One of them is the IDE or the code editor. The other is the command line or the terminal. And of the two, I thought that the relative difference I could make was bigger in the terminal. It's just like the IDE was the more advanced tool, whereas the terminal really hadn't changed in the last 40 years. And the terminal is the thing that if you watch a hacker movie, there's a black screen and the green text and a developer typing in commands. It's like a place you type commands in. It's really like the oldest computing interface, maybe punch cards are older, but it's like the oldest really interactive one. And for a long time, that was how everyone used computers was by typing commands in and getting answers out. Now the world evolved to a world of like graphical user interfaces, and that's been great. But for developers in particular, there's still a ton of utility in using this really text-based input/output oriented interface. It lets you do things more quickly. It's programmable, which is really important. So like every, anything you do in the terminal, you can script and automate. All the programs in the terminal, like, interoperate with each other because they're all text-based. And so it's this super powerful tool that developers use for writing code, for building their code, for deploying it to production. Pretty much all of the servers on the internet, by the way, are just running this text-based code. They don't have graphical user interfaces. And so it's a key tool, but it was something that just really didn't evolve. And there were various reasons for that. I think one reason was like, didn't seem to be any like business incentive to spend a lot of time on the terminal. So what you had was a lot of single developer open source projects making different terminals, which is great, but also I don't think results in the same quality of product as if you have 20 people working full-time on the experience. And so the really big leap of faith that we took when we started the company was not like, is there a better product to be built here? I was very certain that there's a better version of the command line that should exist in the world. But really like, is there a business that justifies raising capital and rebuilding this product to make developers more productive?

I mean, this is a core part of the developer workflow. You're innovating on something that hasn't really changed, and I think it might surprise most people that developers are developing code in the same way that they have for the better half of 30 years, if you will. And so it was time for a change. So in your own words, what is Warp and how does AI supercharge this new terminal experience?

Warp today is a little different than what it was when I started it. Warp today is an interface where developers can just tell their computer what to do and the computer will do it, which is a kind of crazy thing to say, but that could be you tell your computer you want it to code you something. It could be you tell your computer you want it to set up a project, you want help debugging, you want to deploy something to production. It turns out that the command line is actually a great interface for that. And what's actually changed is that when we started Warp, the only way you could tell the computer what to do was by typing commands. And those commands, they're not English, they're cryptic. It's like cd space this or ls git space that. And you basically have a separate language that you had to learn in order to tell your computer what to do. And today you can just tell your computer what to do in English, which is just like mind-boggling as a, as a developer. But that's the power that has been made available through all these advances in LLMs and AI in the last couple years. And this stuff didn't exist when we started the company. When we started the company, we were trying to build the best user interface for running commands and like sharing command knowledge across your team. And now the interface is not just the best interface for running commands, it's really the best interface for telling your computer that you wanted to do something for developers to do that, I would say, because it's still, it's still a technical tool.

I deeply wish I was taking Computer Science 1 in college now because I would probably not have been banging my head against the wall in the same way that I was then trying to teach myself how to code. Claudia actually successfully graduated with a CS degree. I had to do a minor in engineering. That's the best I could do. So I really relate to this problem because the prompt engineering space and of course the whole vibe coding movement really resonates with me as a non-technical PM and having to get up to speed on what's going on in the terminal with my engineers.

It's really wild how much it lowers the barrier to entry. Anyone on our team can use a terminal, which is crazy because it is probably the like most intimidating of any developer tool out there. It's just this blinking cursor, but like this really to a large extent removes the barrier when you can just You can tell it what to do, and if it does something and you don't understand, it will explain it to you. If you hit some error, it will fix it for you. And so it's like you have some terminal genius who's just infinitely patient with you as you're trying to do things in the tool helping you do it. And then it's also like you could just ask that terminal genius to go do a bunch of stuff for you and report back on how far they got, which is cool. So it's totally changing.

If you have not checked out this download Warp now, you can do so for your Mac, as I saw when I was stalking your website earlier. I want to just acknowledge that you've done a Herculean effort of taking a company that had a vision pre the watershed ChatGPT moment, and now are continuing to scale into this AI-native world. So that is quite novel. Tell us about that oh shit moment in November 2022. When this whole thing changed what you were building?

Yeah, this is an awesome question. So it's interesting. So we actually had AI in Warp before ChatGPT. We had something called Codex, which was an early OpenAI API. And so the way it used to work was you could think of it as like English to command translation. So you could be like, what's the command for finding all Rust files in this directory? And it would spit out the find -n, like the whole command. So we had that and then ChatGPT came out and we were like, this is pretty crazy and very applicable to what we're doing. So we were like, what's the fastest thing that we can do to make this useful for our users? And we already had like good amount of users back then. I don't remember how many it was, tens of thousands or maybe 100,000 users or something. And what we did was what a lot of companies did. We made a chat panel. In Warp, you could directly chat with ChatGPT and we're like, oh, this is so cool. You can have it reference things in your terminal session. And we did that really fast. We did it in like a month and it turns out, yeah, that was cool. But the more we thought about it, we were like, it's weird and unnecessary to have like a separate chat interface in the terminal because the primary interface consists of you telling the terminal, like, what you want to do. And chat is not taking advantage of what these models are capable of. Chat, it's very useful, obviously, like ChatGPT is super useful, but the really cool thing is that these models can do things and the terminal is a very natural place for them to do things because the way that you do things is running commands. And so the next thing that we Launch, which is the thing we did an awesome job on, was this thing called Agent Mode, although we're gonna move away from that naming, where you had a choice as a user of either typing a command or just interacting in natural language. And if you interacted in natural language, it wasn't just a chat interaction. It was that the AI could gather information so it could run commands or it could even do tasks. And it would like ask you every time, do you, can I do this? Can I read your files? And so that's where we were like, oh, this is super duper powerful. It was that first hint of like agent. Like interaction, and then we've just progressively leaned into that. And the way that you lean into that is you allow the agent to have more autonomy, you allow the agent to have more context. So figuring out all the different ways that someone's code could get looked at, or what are their READMEs? What's their infrastructure look like? And then finally you give it more tools so that it can do more things. We're at a point in Warp where we launched a beta of this thing a month ago where you can just tell it to do something and then you can go open up some other tab and you can check back and see how it's doing on it. That's the future, is that you don't necessarily as a developer have to like be in this fully interactive mode with the AI at all. You can just tell it to do things. It's cool.

Maybe let us into the room when you were talking to your investors about this need and shift in product direction, because something tells me a lot of companies funded before 2022 are now having a similar aha moment of needing to change the overall product vision?

A couple of cool things. So one is the mission of the company, which you mentioned at the start, was always just let's empower developers to build and ship better software. And so we didn't need to change the mission. It was more like, hey, this will help us achieve that mission. And then the product vision from really early on at Warp, it wasn't like build the world's best terminal. It was to try to build the tool that developers spend their day in shipping software. And so it wasn't like a pivot. It was was like, hey, this is going to let us build the product way more effectively and way better. The other thing that was great about it for us was that it opened up a business model that we didn't have, which was like, the terminal is a piece of client-side software. It's hard to justify charging for something that doesn't have any marginal cost associated with it. But when AI comes along and like dramatically improves a developer's productivity and there's an obvious marginal cost associated with it, it opened up a whole business model that we didn't have, and honestly has been the key to growing our revenue. Our revenue is growing very quick, anywhere like 10, 15% every week this year, which is crazy. In a lot of ways, we just got really lucky, and I guess we did a good job adjusting based on the arrival of the new technology. And then we also had built a lot of the foundational UI, which enabled us to take advantage of it, but some part of it was just luck. Look, I'm being honest.

I'd love to double-click into this sort of growth journey, right? Warp has hundreds of thousands of monthly active users, as you mentioned, like growing 10 to 20% weekly, which is insane. That compounds over time. Tell us about your first growth inflection point a little bit more and really what the core drivers were beyond having an insanely helpful product.

From Selfmade, actually, I should say one of the key things that I took away from that is don't grow before you have retention. Like, you need It's very hard to grow something if you don't have a product that's delivering value and that people want to stick around with. And so very early on, like actually the first thing we did before we launched anything to the public was like, can we dogfood this? Can we use it? Is it actually helpful for us? This was, I'm talking like 2021. We then tried to get friends to dogfood it, and that was kind of a fail. It turns out in our network, we didn't have enough people who randomly wanted to try a new terminal. And so then what we did, we're like, uh-oh, is there not demand for this? And so we did a, a landing page, just like a pure waitlist landing page. And we posted that on Hacker News and we got to top of Hacker News. That was validating. And we got over 10,000 people, 10,000 developers to sign up for the waitlist the first day on Hacker News, which was like, okay, this is great. This is like, people are actually interested in improvement in this product. And then it became, can we get people who are on this waitlist who have indicated they have some sort of demand, can we get them into the product? And so we did a private beta and just followed the retention cohorts. And so at first they were pretty bad, but we got useful feedback and we had a feedback loop going. And then we got the retention to a point where we thought this is actually good enough. And then we did a public launch where we basically ran that whole playbook again of Hacker News, but this time we added on, I think also like a funding announcement and some press and a Discord and like lots of like social stuff. We have this great network of angel investors who put it out and that was enough, it turns out, to get us in the tens of thousands of developers. And then from then out, lots of like word of mouth, people telling each other, and that compounds. We don't have a true network effect. That would be like super game-changing if we could really get the team part of Warp going. We do have a team set of features. I do think there's a lot of value in that. If we could get people, the incremental user using Warp, making their teammates more valuable, that would be a different kind of growth. But we've been able to continue this growth for several years now, just based on a combination of people telling each other, and then we do a little bit of marketing as well to help with it. But most of our users still come from their teammates being like, hey, you got to check out Work, which is incredible.

And I think there's probably a bit of a difference between being in the space prior to the proliferation of AI code assistant tools over the past 6 months with really being a known entity before ChatGPT's watershed moment. And so I think like now there's probably this sort of analysis paralysis of there are so many AI tools, which do we adopt? And so being able to have that core user base prior to a lot of the mania probably has helped.

Another thing that I will say that has helped us and maybe hurt us in some ways too, is that we have a very differentiated product compared to most of the other AI developer tools. In our space, we have companies like Cursor, Windsurf, Copilot. They're all the same product, strangely. They're all VS Code with a little bit of different UIs, but primarily the differentiator is in like the quality of the AI completions, whereas Warp is most definitely not that. Warp is a from-the-ground-up native app with a totally bespoke, unique interface, which is good and bad. The good is I feel like we can do whatever we want with our UX and really build it for these AI workflows in a way that maybe some of these other people in our space can't. The bad is that VS Code Code was already a 70% market share IDE. And so it's like, it's super easy for a developer to like switch from VS Code to Cursor or Windsurf or whatever. You basically get the same experience.

As we think about this scaling journey, we've talked a little bit about how you brought your investors along for the journey. A huge congratulations on fundraising through a $73 million amount in funding with some incredible investors like Sam Altman, Marc Benioff, and then on the VC side, Sequoia and GV. Hallmark names, of course. Who was the first person to say yes to investing in Warp?

There were a couple early investors who were really helpful, so I'll call out Greg Rosen at Box Group and Eric Norlander of GV. GV made sense because I'd been at Google a really long time and actually knew Eric from when I was an engineer, and he was an engineer who'd gone into the investing side, and then Greg was an engineer and he was also an investor in Selfmade, so I knew him from then. And so I've had the very fortunate founder journey here, probably helpful because I'm a second-time founder, of having some of these investors in the very early conversations, even to the point where I was like, should I do this or should I do that? Can you help me talk through the business? What makes the most sense? And when I was ready to be like, hey, I actually want to start this company, which was like 2 weeks before COVID It was really crazy time. They were like, okay, we're gonna, we're gonna back you. And I think they backed me, I think because they knew me, they liked me, they liked the space. So I had backing out of the gate, which was, I don't know, that's lucky too. I think, I think I was very fortunate there. And then all of our funding rounds have actually grown out of the past funding round. So I brought in angels into the seed round, and one of those angels was Dylan Field, who's the Figma founder who ended up leading the Series A. and then Dylan introduced me to Sequoia and they ended up leading the Series B and they did it very early on. And so I haven't had to go out and do like a round of pitching 20 VCs for Warp. I did have to do that for Selfmade and that's not super fun. So I was, I think I'm really fortunate in that respect.

I think probably the second time you go around and the fundraising is not 50 conversations, you realize what it feels like for a fundraise to go well and not so well. But you have such a successful time fundraising, this great organic growth momentum on the product, all incredible milestones in an early startup journey. But as we all know, it's not always up into the right. The founder journey's full of high highs and low lows. What's an unexpected challenge that you faced or something that you went through in the early days where things just didn't go according to plan?

God, startups are so hard. They really are. So, we took a lot of shit on Hacker News. Because we had a login requirement for Warp. And it's like an offensive thing to a lot of developers to have login in the terminal, which by the way, we've, we got rid of it. I was very reluctant to get rid of it. My feeling was like, kind of who cares? Like I log into every single cloud thing that I use, but developers feel like it's like an undue burden or invasion of privacy or just what are you doing? And so for a long time. I think that cost us on the brand side and it cost us on the growth side. And I was kind of obstinate about it and wish I had, you know, been like a little bit more flexible and a little bit like less opinionated on it and listened to users a bit more. And I'm glad that we ended up making the change to do what they want, but I didn't know that was going to be such a thing. But man, like, I don't know if you spend any time on Hacker News, it's a tough crowd and we have like we have like, you know, we've done really well there, but we're also very controversial because of something silly like that. So I don't know, I, I didn't foresee that, but I should have.

As we take a step back and think through these moments of lows and highs and everything that's come about, I think one area it feels like you guys are in, in terms of inflection point, is really figuring out like the scalable business model, right? And go-to-market strategy. And we talk about the this growth that's happening, but it sounds like there's this individual contributor who's excited to use Warp and there's a ton of different pricing options on your website, which feel more product-led growth, go-to-market focus. Then I'm over here, I'm seeing the enterprise tab. So could you maybe bring our listeners into that conversation that sounds like you might be actively discussing with your leadership team on how to position and grow into this dual approach of PLG and enterprise?

Great question. Yeah, the primary go-to-market for us has always been PLG and bottoms-up adoption. For a developer tool, developers, I think, are very opinionated and very influential. And to try to do something where you sell to like a VP and they're like, hey, now everyone's switching their terminal to use Warp, is just a non-starter. And so we realized we had to build something that developers love and want to use. And that has been most of our focus. That's true even on the revenue side. Our focus is more on a self-serve revenue. So like individuals paying for pro plans or small teams paying for plans, but it is all one funnel. And the more individuals and small teams we get paying us or even just using Warp, the more enterprise opportunities it opens up because What will happen is, you know, an individual or a tech lead or kind of a engineering manager will be like, hey, this is a super valuable tool. I think it would be useful for our company to have this be generally available for more developers to use. And that is the source of most of our enterprise motion. We have a bunch of really good enterprise customers as well. I don't exactly know who I have like rights to say, but we have enterprise customers, but it almost always comes from individual developers who are like, this is a power-up. That's the motion. I do think eventually we will be mostly an enterprise business if we execute well, just because that's like where there's more money and like having like individual developers creates other problems. They're more likely to have revenue churn, but bottoms up is definitely the motion for us.

It's nice to see that the PLG approach is working in AI-native companies. And I think that's a bit of an existential question at the moment. Do we need new business models to support what is happening on the ground in the technology?

I don't exactly know what's going to happen. I think for a lot of the AI companies, big questions around revenue retention, big questions around like unit economics. Like these are more expensive services to provide that have like real COGS compared to like your typical SaaS business where by default the margins are really high. You're just paying for database rows and like a little bit of compute. These are pretty expensive businesses to run. And it's really interesting to me, like, who's going to end up making money off of this? Because there's a lot of value for sure. Is it going to be at the application layers? That's our bet. That's where our company is. Is it going to be at the model layer? Is it going to be the Googles and AWSs that make money? Is it going to be just the chip makers? I don't really know. It's a super interesting question.

We can pontificate all about that all day long because I certainly am enjoying these conversations. I'm actually really curious to get your perspective on the competitive suite, and we've a little bit about some other companies, but I'm gonna lay out a framework. You tell me if this lands. So, I'm thinking about the Lovables of the world, the Replets of the world, which are really having their moment in the vibe coding movement around making code accessible, not scary, not the black terminal. And we are seeing this bringing this onslaught of new kind of coders, new engineers. Then we have really, I think, the world that you're living in, which is much more the real developers, and not to throw shade on the vibe coders, but people who were trained software engineers, and this is their day job, they're spending 8 hours a day in a terminal. And that's maybe more the Windsors of the world, GitHub Copilot, et cetera. Does that sound right? Am I framing that up?

That's a correct division. So like Replit, Lovable, Bolt are prompt-to-app things. And I think that their strongest product market fit is actually not with pro developers. It's with people who maybe didn't know how to develop or just want to develop prototypes of things. And Warp is very much probably the most extreme on the like for pro developers because of where the tool sits, you know, what it is.

Where are these worlds converging?

If this is creating a larger TAM for developers overall, or people who think they can be developers, are you going to be going more into the prosumer, or are they going to be going more into the developer?

It's a great question. So we don't think of, at least right now, the Bullets, Lovables, Replit are not our primary competitor that we're concerned with. We're much more concerned with the Cursor, Copilot, Windsurf, Cloud Code is another one that's like really doing something similar to us. My take on this is still, and I don't know how long this will be true for, like the jobs that professional developers are doing are different than what's in the sweet spot for Bolt and Lovable and those. If you're like, you're working at Amazon or you're working at a bank or you're working at an e-commerce company or whatever, you're working on some big existing code base. There's like correctness requirements that it's not about going from like 0 to 1 nearly as much as it is from making changes, incremental changes to a big existing thing that has a lot of customers that might have like real strict requirements. I think that's like a sort of different ballgame than the 0 to 1, build me an app from a prompt type thing. I'm curious if they will try to go like further, like down the stack to where we are. They might. We also in Warp and in Cursor and Windsurf, you can do all of these zero to one things as well. So I, I don't really know. I think whoever's going to win is going to be whomever builds the best interface for telling your computer what to do and making it have properties of like best autonomy, best at getting the context, best at looping you in and giving you a fallback when you need a real developer. And then whoever can figure out what makes that sticky, I think is the other really big thing in our space because like I said, like Cursor, Windsurf, Copilot, I could literally have all 3 of those open and be working in them at the same time. And there's no switching costs, there's no network effect. It's not super clear to me who is going to be able to like actually long-term capture all the value, cuz there's a lot of people working at it. I mean, we're trying to—

we're excited to figure it out with you.

Yeah.

We'll have to have you come back on when this story has been unfolding a bit more. Just a hot off the press question as we wrap up here shortly. What is your quick take on Model Context Protocol?

I'm bullish long-term. It's overhyped in the short term. It's not that useful, but we're adding support for it in Warp and I think it's good.

Awesome.

I love it. You're adding some context there, pun intended. What else is new on the roadmap? What can anyone who's in the Warp space or thinking about becoming a part of your community be expecting from a feature release perspective?

You can check out our preview build. This is something that we launched last week where if you want early access to features, you can get it. It's just on our website. We are going to be doing a major expansion of our functionality where we are less and less just confining ourselves to being a terminal and more and more going for the big product vision that we have, which is that Warp should be an app that you can do everything as a developer in. That's coming. We're going to have a big launch around that in a not too distant future. We'll call it Q2 where It's going to be a really big moment for the company and for all of our existing users. I think it's going to really open the door to doing more types of things directly in Work.

And are you hiring?

We are hiring. Yeah, we're hiring engineers. We're hiring on the data side. We are hiring our first person on the sales side to handle some of the inbound interest that we're getting. So we're definitely hiring.

We'll make a shout out for that too. All right, Zach, this has been such a fun conversation. Thank you so much for letting us into your story and founder journey. We just have our hero question left. Who is a woman in your life that has had a profound impact on you and your career?

I'm gonna go with Shanna Fisher, who is a, a VC in New York who was— when I left Google, I had no concept at all of what would be required to be an entrepreneur, and she was one of the first people who I was introduced to. And she helped me get my head straight in terms of what's required to be founding a company. She really helped me a lot when I was at Selfmade, my first company, and then has been an investor in Warp. And she's just an awesome person for early-stage founders to work with. So shout out to Shanna.

Shanna, thanks for all the support. Glad that Zach's here in the room with us today and just generally so grateful that we got the chance to tell your incredible story. And thanks so much for taking the time with us.

Us.

Okay. Thanks so much for having Madison. This was awesome.

Thank you so much for joining us at The Room Podcast. If you want more from The Room every week, subscribe to our newsletter at theroompodcast.com/newsletter. We'll be back next week with a new episode, an inspirational guest, Tuesday, 10:00 AM Eastern, 7:00 AM Pacific. See you in the room. Perkins Coie supports the most innovative entrepreneurs and investors in fast-moving and high-growth sectors. Addressing their myriad of legal needs. But the firm doesn't just provide end-to-end legal and business counseling to its startup clients. It also facilitates introductions to key advisors and sources of capital. Perkins Coie's interactive website, Startup Percolator, offers access to programs, resources, and rich dynamic content designed to assist entrepreneurs on their startup journey. To learn more, go to startuppercolator.com. And perkinscoie.com.

This podcast is brought to you by Mercury, the banking platform businesses like The Room Podcast use to simplify their finances. It's time banking did more than hold your money. Now it can. With Mercury, you can pay bills in seconds, close the books faster, and even send invoices. Not only does Mercury do away with a patchwork of tools, it eliminates guesswork Giving you complete and accurate visibility into your business's finances, all from one account. Apply in minutes at mercury.com.