My First Million
My First Million

How I built a $3B AI Startup + 7 AI Business Ideas

Episode 711: Shaan Puri ( https://x.com/ShaanVP ) talks to Guillermo Rauch ( https://x.com/rauchg ) about building Vercel plus 7 business ideas to start in AI. — Show Notes: (0:00) Dropout to job offer from Facebook (3:38) Genius vs accident (21:04) Building Vercel (25:26) AI startup ideas (31:38) I

Featured Speakers

Sam Parr & Shaan Puri Host

Topics Discussed

Episode Summary

Executive Summary: The conversation traces Guillermo Rauch’s path from an Argentina childhood shaped by economic instability to building Vercel and v0 around a single obsession: making software creation and deployment radically faster. He argues that open source, JavaScript’s unfair browser advantage, and AI-driven “software 2.0” all reward the same principle—reduce friction, maximize iteration speed, and use narrow, opinionated workflows to let more people build.

Main Topics: From Argentina to open source success (Priority: 5/5): Rauch explains how early coding, teaching tutorials, and open source contributions built his career from age 10 through teenage years, culminating in job opportunities and leaving high school for work in Switzerland. JavaScript as an unfair advantage (Priority: 5/5): He describes betting early on JavaScript because browsers could run only one language, and argues its success came from constraint, evolution, and the ability to bootstrap richer capabilities over time. Vercel’s founding principle: minimize time from idea to URL (Priority: 5/5): Rauch says Vercel was built to remove deployment friction and optimize iteration velocity, so teams can ship and test ideas in seconds rather than hours or weeks. v0 and AI as the next leap in creation (Priority: 5/5): He frames v0 as ChatGPT for building web apps and says AI will let users go from prompt to application while still relying on strong software 1.0 infrastructure underneath. Future verticals: video games, forms, e-commerce, legal tools (Priority: 4/5): Rauch predicts AI-first platforms will emerge in many domains by unbundling common ChatGPT use cases into specialized products, with opinionated workflows and model orchestration. Bitcoin, Argentina, and trust in money (Priority: 4/5): He connects his enthusiasm for Bitcoin to Argentina’s currency crises and argues that immutable, cryptographically verifiable money is a necessary response to institutional failure. Ideas from the 'free AI ideas' thread (Priority: 4/5): He riffs on product concepts such as AI form builders, AI camera/photo editing, and absurdly smart autocomplete, emphasizing context-aware interfaces over rigid forms and menus.

Key Arguments: Open source was the main lever of Rauch’s early career, because it created reputation, proof of skill, and career opportunities before traditional credentials mattered. JavaScript won not because it was initially the best language, but because it had the browser monopoly and could evolve from a minimal, constrained starting point. Successful products often begin rough and narrow; looking at early websites and prototypes reveals that simplicity and speed matter more than polish at launch. Iteration velocity is the core optimization: the best tooling shortens the path from idea to shared artifact, and Vercel was designed around that principle. v0 extends the same philosophy to AI: instead of generating text, it generates working web applications while leaning on Vercel’s production infrastructure. The next wave of AI startups will not train foundation models from scratch; they will combine existing models into opinionated workflows for specific tasks and industries. Many AI opportunities are hidden inside common ChatGPT query patterns, which can be turned into standalone products like legal drafting, form creation, or game creation. Bitcoin resonated with Rauch because Argentina taught him that governments and banks can fail catastrophically, so he values money with strong trust guarantees and immutability.

Data Points: Age started coding seriously: 10 years old - Rauch says he began creating websites and coding seriously at about age 10. First freelance check: 11 years old - He recalls getting his first payment while still a child. First client abroad: Netherlands at 12–13 - He mentions an early client in the Netherlands before adulthood. High school entry rank: #10 out of thousands - He says he entered a selective public school in Argentina in position number 10. JavaScript language market advantage: 1 browser language - He argues browsers can run JavaScript but not Python, C++, or Java on the client side. MooTools contribution age: 15–16 years old - He says his big break came from contributing to MooTools as a teenager. Facebook job offer age: ~16–17 years old - He says Facebook made an early offer, but it fell through because he was in Argentina and underage. Time to conceive JavaScript: 10 days - He cites Brendan Eich’s timeline for the language’s creation under deadline pressure. Vercel deployment goal: Seconds - He says Vercel was built to reduce idea-to-live deployment from hours/weeks to seconds. Vercel’s scale claim: $3 billion valuation - The interviewer references Vercel’s approximate valuation during the intro. Conversion lift from site speed: 30% to 90% - Rauch cites a public company that improved conversion by speeding up its website. Coinbase BTC promo value at the time: $10.81 - He shows a 2012 email offering Bitcoin for installing the app. Coinbase BTC promo value now: $103,000 - He says the same Bitcoin would be worth about $103,000 as of that morning. Bitcoin email date: November 9, 2012 - Shown on the printed Coinbase email exchange.

Pivotal Quotes: "I no longer write code. I only prompt." — Guillermo Rauch: He describes how AI has changed his engineering workflow. "The most important thing is to get the idea out into the world as a URL and see if it sticks." — Guillermo Rauch: He explains the Vercel philosophy of minimizing friction from concept to live product. "You cannot trust any given actor. You cannot trust the government. You cannot trust the banks. You cannot trust your friends. You cannot trust anyone." — Guillermo Rauch: He explains why Argentina pushed him toward Bitcoin as a trust-minimizing monetary system.

Implications: The episode suggests the next generation of software will be built by combining AI with opinionated infrastructure, not by replacing it. For founders, the edge is still speed, context, and workflow design; for users, creation is becoming accessible through natural language.

🔓 Sign Up for Unlimited Episode Search

About My First Million

Sam Parr and Shaan Puri brainstorm new business ideas based on trends & opportunities they see in the market. Sometimes they bring on famous guests to brainstorm with them.

View all episodes from My First Million