Guillermo Rauch: The Product Visionary Fueling Vercel and Millions of Developers

0:00This last week, we had three customers doing ads on the Super Bowl. Guillermo, seven years ago, would be petrified because he would be worried that the customers were responsible for that code. And it would be basically impossible to get 100 million people watching an ad going to your website and scaling. Guillermo today, developed Next.js, almost guarantees certain outcomes that seem nearly impossible to do yourself. That's the voice of Guillermo Rauch, founder and CEO of Vercel. His blend of technical foresight and audacious ambition, coupled with an obsessive attention to detail and products, has turned what was once a humble deployment tool into an enterprise-scale platform, handling even the massive traffic spikes of Super Bowl ads.

0:49This is Mike Maples Jr. of Floodgate, and it's Go Time with Guillermo Rauch. This is Mike Maples Jr., and welcome to the Pattern Breakers podcast, where we explore why some founders radically change the future and how they stand apart. Together we'll learn about the counterintuitive mindsets and actions behind their remarkable success. Brace yourself for a world where chaos is welcome, naysayers are often a positive signal, and movements galvanize misfits who transform the impossible to the inevitable.

1:32Vercel is used by over a million developers each month, from small companies to tech giants like Netflix and Scale AI who want to create fast and reliable websites. But in addition to Vercel's success today, much of what I admire about Guillermo's journey is that it spans decades. Even before ideas like serverless computing and AI-powered apps became popular, Guillermo was already working on open source projects for the modern web. As a teenager in Argentina, he built an early tool to automate WordPress tasks, contributed to projects that changed web development, and later helped launch Next.js, a tool now used by many leading companies throughout the world.

2:16He makes smart choices about what to build next because he's been thinking about these challenges all the time, for years or even decades at a time. When he acts, he moves quickly while balancing big ideas with practical details. He also has a talent for storytelling, a love for learning, and a deep curiosity about the major changes in AI and the future of the web. Guillermo serves as a great example for founders who want to build enduring technology companies from the ground up. Let's talk to him. Guillermo, welcome to the podcast. Thanks for having me, Mike. I've been looking forward to this.

2:57You've been coding for a while. You're a young guy, but you've been doing this for some time. So when did you first start using computers and learning how to code? Using computers at our home in a little neighborhood in Argentina was kind of an interesting adventure. We got a PC for the house and my dad's first challenge for me and my siblings was, can you do something cool? Can you hack around its limits? Can we get some new software on it? So it was kind of like a love at first sight to try and get the most juice possible out of computers. And later on, I started this journey of programming and I had a bunch of false starts, different programming languages that I tried, but I really fell in love with the web.

3:42And I kind of fell in love with like systems administration with Linux and programming for the web with PHP and technologies like that. So open source has been my entire life, really. And it was the key thing that in many ways created a career out of a very young person in Argentina. It was contributing to open source and getting to know all these amazing people all over the planet that were also passionate about the web, open source, and creating new things and trying to make the web a better place. And so when did you make your first open source contribution? What age were you, do you think?

4:17You know, actually, the first kind of contribution I think is really important and underrated is whenever I would learn something, I would go and teach it. I didn't know at the time that's the best way to learn anything, right? Like the very first thing, if you can synthesize that knowledge and really assimilate it, is a Richard Feynman thing, right? When you can explain it simply, you truly understand it. I would learn something about Linux, and then I would go to forums and community chats. And then if someone had the same problem I had, I would try to help them. And that wasn't quite open source in the sense of like, I'm contributing code.

4:52But a lot of people now come to me and say, hey, I love your projects, Next.js. I love Turbo. I love the AISDK. How can I contribute? And in a lot of cases, I tell them, teach. And then it probably came in the form of starting to create plugins for existing systems. Another really good way of getting into open source. Creating a whole project and platform is hard. It can be daunting. but a lot of systems have plugin mechanisms. And you can actually create an enormous amount of value and leverage by creating those plugins. I can't remember the specific first thing, but I'll tell you a fun anecdote.

5:27People are talking a lot about generative SEO and creating content on autopilot. One of my first breakthroughs was I created a WordPress plugin that generated content. We didn't have LLMs at the time, but you could use RSS. to actually get dynamic data. And you had little like templating features into the, I called it WP-O-Matic. It was actually a huge breakthrough for me because it taught me that people were very interested in getting professional support, that they could buy it. I got a lot of leads. People wanted to acquire my plugin and I was like, you know, 12, 13 years old at the time. So it was a fun time.

6:07And do you think that growing up in Argentina, did it change your perspective, you As a person now in Silicon Valley, do you think that that kind of outsider's perspective? It's huge. It's huge. Culturally, the thing that really impacted me when I first came to Silicon Valley was that you can just build things. There's an appetite for innovation and an open arms culture for anyone that wants to create something great and has a big, bold vision of the future that I didn't feel like we had in Argentina. Now I'm a lot more optimistic now because of the new governments. There's a vibrant startup ecosystem growing.

6:49There's more narrative, not around the specific policies, but like the pursuit of freedom and innovation and appreciation for people that build things that is now nascent in Argentina. You've probably seen Millay and Elon hanging out. That appreciation of people that want to push the frontier of human attainment, I didn't grow up with directly. I grew up because my parents had a lot of appreciation for the American system. My mom worked reselling instrumentation hardware for a company called Fisher Rosemount. She appreciated that the quality of documentation, the quality of a specification, the complete product was better.

7:28So I grew up with, oh, where can I go and create things that are more complete, higher quality, better? And how can I surround myself with people that appreciate those things as well? And so how old were you when you came to Silicon Valley then? So the first visit was probably when I was 18. So I was telling you about like, I did things with WordPress. I did things with forums. I had my sort of contributions in the form of teaching. And I had some really good blog posts. But the big breakthrough for me was this project, Mood Tools, because it started having a lot of impact in the world. It was a library for user interfaces.

8:04But MoodTools kind of struck a chord with the community and startups began to choose it. Facebook chose our library. So Facebook was using it a little bit, but also decided, hey, like this MoodTools team shares a lot of our philosophies and they ended up hiring a bunch of those people. And so I always tell the anecdote of a Facebook recruiter reached out to me in the early days. Well, they turned me down because I was 17 and in Argentina and they were trying to hire senior engineers in Silicon Valley, right? But having that exposure to open source opened a lot of doors for me. And it was kind of love at first sight for all the reasons that I mentioned, where as soon as I arrived here, I felt like this was the best place in the world to create new things.

8:45The first meetup that I went to, people immediately knew me. Like no one knew me anywhere. But in that meetup, people were like, wait, Mootools? Like, oh, I've used your thing. And that actually was kind of one of the first signals. Like, wait, people really appreciate tools and infrastructure. Imagine a place that appreciates things that are not even visible. And for me, being in that world of designing open source libraries and frameworks, I actually found like person location fit. So you've been thinking about rich web applications for a while. In fact, you kind of have an evergreen blog post on principles of rich web applications where you laid out some fundamental ideas.

9:29Have those principles evolved as you built Vercel or is Vercel kind of just actualizing them one by one to the best degree that you can? It's so funny you call that out because there's a specific times in history where some essays need to be written. I put so much energy and effort into trying to create the most accurate prediction of the future that I could. What would it look like if the entire world wanted to say, I'm going to have the best possible website or web application? What are the constituent factors? What are the things that these applications should do? that essay became the foundation for Vercel and Next.js because I realized first seven principles is a lot and also the explanation of each principle could have been a book yep and I went on this conference tour so at the time and this actually helped me get my o1 visa you were you were asking like when did you first come here well I'm a high school dropout and I had no no no accreditation of any kind, no college degree, obviously, no very little job experience.

10:36What I did have in spades was extremely influential open source projects. And I was getting invited to speak about my projects and these principles all over the world. And I remember giving this talk about the seven principles of Ridgeway Applications. And I realized, well, giving that talk, I could just create a technology that embeds these principles and puts them on autopilot for the world. I don't have to go to every city in the world and get thousands of engineers in a room and tell them, look, you have to do pre-fetching. You have to predict the future. You have to use server-side rendering and pre-rendering so that you optimize how much work is done in the cloud instead of the device.

11:17And it didn't make the direct connection, you know, like, oh, I'm going to go and create a technology that replaces this essay. but it kind of happened serendipitously in some ways. And I'm very thankful that I have that to look back on and it's held amazingly well. The framework that I created Next.js has now 1.3 million monthly active developers and it powers a huge percentage of the web, especially the newer applications that are being created, the revolutionary AI applications like 11 Labs and Perplexity and MidJourney. They're built on this open source framework. and I could have never imagined that it was ever going to get that big.

11:57But what I did see early on was that when Facebook open source React, there was something very special about that technology. And I wrote an essay about it because it was fundamentally changing how people wrote code. It went from imperative to functional programming language. It went from a mechanism that allowed applications that were very buggy to be shipped to a new way of even interpreting the world that made applications more robust, more ambitious, more real time. And I fell in love with that technology. And I today encourage entrepreneurs to ride the right waves, I call it. When React came out, I tell the story of, I don't know any person that rode that wave early on and today are not doing amazing things or had an amazing outcome from understanding the potential, and betting on the potential of a technology like that in the early days.

12:51Yeah. So were those the main things? So there was just your own experience, which led you to Next.js, and then there was React. Were there any other major market or technology shifts that basically created the inflections for Vercel? The trend that I wrote with Vercel that ultimately helped us win that initial battle for the future of the web is the move from static to dynamic experiences. And there were proof points along the way of applications like Gmail and Facebook News Feed and Facebook Chat and things like that that were full-blown applications that were taking over the world. And I wanted to create a framework that didn't put the web in a little box and said the web is for documents.

13:40I have a lot of respect and admiration for Apple and I'm also very critical of how they treat the web. In their mind, Apple ships web browsers that say the web is for Wikipedia. It's for static documents that you swipe left and right like a little book. Like even their animation is like you're flipping pages in a book. And for me, the metaphor that I give people is like, no, no, it's like it's alive. It's more like the Harry Potter newspaper that like they open it and like magic emerges from it and everything comes alive. That is the web. And it's actually very, I'm not going to say very easy.

14:16out of respect for actually the challenges of running internet services, but it's much easier to create cloud infrastructure to host the static pages than the journey that Vercel and myself took on. Basically going into mortar and figuring out how to scale computation for arbitrary workloads for millions of signups on our platform at all levels of scale to render dynamic pages, personalized experiences, and ultimately AI native experiences. So I think we placed the right bet on the web was not going to be static, but it came with some short-term pains and challenges for sure. Yeah. And it wasn't obvious, right?

14:57Because for a while, people were saying HTML5 was going to be the answer, right? And most of the people who had their romance with HTML5 kind of got burned, right? Yeah. I've always loved systems that scale up and down and systems that scale in and out. You need to be able to create MikeMaplesJr.com and it says, hello world, I'm Mike. This is my latest photo. These are the five things that I'm working on and it's just a static page. Next.js can do that. But where we didn't stop is that if you say, look, I also want to create ChatGPT, which is created with Next.js and it's going to be completely dynamic.

15:34It's going to have login and logged out state. Every time you type in something, it's going to transform the UI dynamically on the client side and it's going to stream tokens from an AI and it's going to convert them to user interfaces just in time. The same framework can also do that. And so I thought very hard about, can I create the same tool that progressively discloses complexity to the end user and you don't have to outgrow it? And lo and behold, now Next.js powers Walmart and Starbucks and Goldman Sachs and McDonald's and a lot of the AI companies and also little personal websites. So the key thing was to have a respect for HTML as a foundational property, but also give people the tools that also push the web forward.

16:25And so when you started Vercel, did it take off right away or did it take a while? What were the early days like and how was it in the early innings? And so when I first put out Vercel, the idea was, I'm going to make it so easy to deploy a website or web application that people will forget AWS even exists. No one will have to configure anything ever again. You're just going to focus on creating your web application and you're going to run one command and it's going to be instantaneously in the cloud. and developers in Silicon Valley and throughout the world immediately said, oh my God, like if you make this happen and if you continue to build this out, this will completely transform how I use the cloud.

17:10But making that system scale and be best in class was the journey that followed in years to come. And it was a very difficult journey. So when I came to market, I thought that I was going to win because I was going to make all software on the planet, All of it instantly deployable with one command. And now I've turned this into a principle. You have to balance infinite ambition with respect and realism about the product surface that you take on in order to eventually get to that ambition. And so the magic really happened as the journey continued when we created Next.js because we saw an opportunity to refine the deployment surface to frameworks and applications like Next.js, as opposed to saying, I'm going to deploy a bank-grade COBOL application on Vercel.

18:06And at that time, I thought, well, all software should be deployable, right? So maybe it can take on every single language, runtime, framework at the same time. And so it took a lot of refining and narrowing the focus in order for us to really uncover this massive opportunity. So we refined our focus from being everything to being a front-end focused cloud. And that actually resonated extremely well with the market. I will argue that it was almost hard to discover who our ideal customer was because the web is so versatile that everybody can use the web to build and deploy anything, right? But what I needed to find was something to excel in.

18:47I needed to be world-class at something so that I could survive evaluation processes where like the champion gets excited, presents it to the boss, the boss gets excited, and then it finally lands in production. And the journey doesn't even end up there, right? You have to do a great job in production for the customer. You have to produce a return on their business legitimately. With the front end focus, we're able to do that because we started telling the story of like, if you make your front end faster, you'll make more money. You move to Purcell, it's faster, you make more money or you get more signups.

19:20Of course, took a lot of pain, took a lot of shifts, took a lot of adapting the infrastructure, perfecting it, replacing parts, learning about how to be secure internet service that's constantly under attack and mitigating cyber threats. And so the journey is sort of endless in sort of, like I mentioned, perfecting that focus, but that's really what unlocked a lot of growth for us. And so was that kind of the, when you started to achieve the breakthrough, it's kind of like when you decided to narrow down to the developer, the end user developer experience and just crush that, you know, stick the landing in that topic.

19:57Yeah, it was this convergence of things. There was a shift from backend to frontend. There was the birth of React. Enterprises were starting to get really interested in how to adopt React. But the biggest unlock, which is something that I think is something that a lot of entrepreneurs are now beginning to get inspired from, it's almost like a principle, is this idea of framework-defined infrastructure. We're not a cloud platform where you sort of assemble every piece by hand. We kind of automated the cloud away by focusing on these frameworks. So the real breakthrough was that Next.js started to take off.

20:33companies really strongly related to this front-end focused way of building. The developer excitement was absolutely enormous, only comparable in my career to what we're seeing now with AI. And this premise of the cloud is extremely enticing, but it's so hard. I talk to CIOs and CTOs still who have been trying to get to the cloud for a decade. And sometimes in the old way of doing things, they didn't see the value. Think about your HR system. Think about any software that you don't like. And think about the cloud before Vercel, right? And others like us. You were supposed to take that software that people hate and you put it in the cloud.

21:15And how does it get better exactly? Instead of paying for the data center, I pay for the cloud data center. And even the abstractions are comparable, which is actually hugely disappointing because the abstraction that I deal with in data center is computers. And I'm in charge of putting the software in them and managing them. And when we went to the cloud, same thing, sounds like a scam. I have virtual computers now and I have to manage how to put the software and make it run smoothly. And so the biggest unlock is that Vercel said, look, it's going to run better. If you do take the journey towards us, it's going to be easy.

21:49You're going to adopt Next.js or other 34 frameworks that we support. You're going to deploy and you're going to make the most efficient use of the cloud possible that makes your business better. Who were the early customers who helped you figure this out? It's kind of like Brian Chesky tells the story about he goes to New York and he sees the Airbnb host and he's got a spiral bound notebook, basically better roadmap than what he had. Yes. Any customers like that? One of the things that I noticed is that if people were left to their own devices and needed to figure out the frameworks and tools themselves, they didn't always accomplish good long-term outcomes.

22:27The website would be slightly slow. Maybe it would crash. Or maybe they would get on the front page of Hacker News or Reddit or whatever, and the website would crash. And I started realizing that's also on me. This last week, we had three customers doing ads on the Super Bowl. Guillermo, seven years ago, would be petrified because he would be worried that the customers were responsible for that code. and it would be basically impossible to get 100 million people watching an ad going to your website and scaling. Guillermo today developed Next.js, thought about all the guardrails and optimizations that when deployed to our cloud, which is an option, Next.js can be deployed into many clouds, almost guarantee certain outcomes that seem nearly impossible to do yourself.

23:17One of our customers, Ramp, came up with the idea of doing the advertising campaign with a football player. And they had to put together an entire campaign in two weeks. I don't think there is any infrastructure on the planet that without any custom configuration can receive a 43X growth in volume in a matter of seconds. Meaning I give a URL out into the world and immediately gets 43 times the baseline load. And so we designed a system that assumes that end-to-end ownership and basically aligns the code that the developers write and increasingly that the AI is right with those exceptional end-user outcomes.

23:59Some of the founders I work with can sometimes struggle with the story, how to tell the story about their startup. Did you find that you just came by that skill pretty naturally? No, I think you work on it, like you work on everything, even aesthetic sense, by the way. Like I worked extremely hard in refining my eye for what is a good user interface. And there's a visual alignment formula as well. It's not just mathematical purely. It's what makes things look good. The weight of typefaces, the distribution of objects in space and how much blank space is between these objects. So you can train your eye for what good design is.

24:41To some people, it comes more naturally. And to some people, a lot of exposure to good design and what people consider to be good design helps a lot. And I think the same is true for storytelling. I think being able to put yourself out there like I did in front of a lot of stages of people and seeing what resonates and seeing what doesn't resonate. And I think that's a little bit of a lost art with entrepreneurs to your point. I'm glad that you're coaching them. It's a good reminder for me. I love the quote from Don Valentine, the founder of Sequoia, that the money flows as a function of the stories.

25:22You're telling a story of the world. You're inspiring them. You're telling stories to investors. You're inspiring them. And that's how you build the following of who your customers will be and for recruiting the best people on the planet, storytelling. But you seem to have that you seem to come by it pretty naturally in terms of just, you know, in many ways, a story, a great story moves people. And on some level, a story is a movement, right? They're kind of related in some ways. And it seems that perhaps by coming up through open source, you know, your early influencers. uh there's that and and there's also and maybe this is good advice for other entrepreneurs is that people really if it works for you people really appreciate your own story you know for me i was always obsessed with the web and front engineering and open source and so the fact that i'm creating a company that tells that story even to a much broader audience and continues to invest in that direction is more genuine than someone that just stumbles upon a hot idea.

26:30And people might think, well, if they just stumbled upon this idea, will they be doing it 10 years from now, 20 years from now, 50 years from now? And I think this is why these love that those obsessive personalities that people have just been grinding on this thought for so long. And every successful technology that I'm a huge fan of has come out this way, right? You look at the origins of Bitcoin and the people that were working on this mailing list, the cyberpunks mailing list. Bitcoin didn't just come out of nowhere. It came out of people just being obsessed with the problem space in ways that weren't popular or sexy for a long time.

27:09That's true to me. JavaScript was a horrible laughingstock of the programming world. In fact, as the industry matured and noticed the value of the end user experience, did people start taking it more seriously? Because as I mentioned, it started being correlated with conversion and business value and customer loyalty and retention. Like it took a lot of me investing in that world before it was cool in many ways. And so I think having that long-term mindset helps a lot in overcoming the naysayers that come here and there. Yeah. And the other thing I've seen is this is not just true in the tech entrepreneurship field.

27:52Scientists, artists, they describe the same type of thing. Somebody supposedly asked Newton at a party, what was it about the apple falling on your head that caused you to realize gravity? And he said, because I thought about it all the time. you know that's so true right like or louis pasteur chance favors a prepared mind well prepared minds are more likely to get lucky because they they answer the door when the muse knocks uh because they're not trying to just knock through their to-do list tasks of the day they're sort of pursuing something that they're obsessively interested in i do a thing like that that I love, which is I think in the background about a problem, sometimes for months, sometimes for years.