Garry Tan: Own Your Intelligence
Y Combinator · 2026-08-06 · 42 min
https://www.youtube.com/watch?v=eRrc1pUY5oUVideo summary
Garry Tan argues personal AGI, G Brain, and Markdown skills let one founder achieve 400x leverage while owning their intelligence.
At Startup School 2026, Garry Tan frames personal AGI as intelligence you own rather than rent: an agent running on your infrastructure, using a personal library, and executing Markdown-based skills that compound over time. He cites his own results—about 400x more coding output than in 2013, still at least 8x after harsh adjustments—and says a quarter of YC’s Winter 2025 companies had codebases that were 95% AI-generated. Tan’s G Brain contains roughly 220,000 Markdown pages covering 25 years of life, while its open-source G stack has 123,000 GitHub stars. His five-step plan is to run a local agent, build a small personal library, write one skill file, automate it on a schedule, and convert every task into reusable memory. The core warning: if your skills live in an employer’s repository, your career becomes extraction; if you control them, you operate under your own power.
Chapters
- 0:00What Founders Can Learn From Spinoza: Conatus and Building Without Permission
- 4:46Personal AGI Is Already Here: Intelligence You Own Instead of Rent
- 7:57Why AI Makes One Person More Powerful Than Ever: 400x Output and Context Leverage
- 12:29Your Life Is a Library: Seven Items Versus One Million Tokens
- 15:15Inside My Personal AI System: G Stack and Context-Aware Daily Work
- 17:06Markdown Is Code: Skill Files Turn Everyone Into Agent Managers
- 18:20Latent Space vs. Deterministic Code: Pairing Judgment With Databases
- 21:13Building a Company of One: Agent Organizations and Tiny Teams
- 24:16How to Build Your Own Personal AGI: Five Steps to a Compounding System
- 27:58Why Most People Will Quit Too Soon: The Week-12 Compounding Curve
- 29:09Own Your Skills Before Someone Else Does: Maya’s 40-File Career Choice
- 32:24Personal AGI Means Owning Your Intelligence: Staying Under Your Own Power
- 35:21Why I Open Sourced Everything: Giving Away the Markdown Workforce
- 38:30A Personal AGI for One Small Boy: 80,000 Files for Epilepsy Research
- 40:18It's All Made Up. You Get to Make It Up.: 7,000 Founders with Direct Leverage
This is a Tier 1 public summary
Whether the chapter key points, section summaries and mind map are public is up to the person who shared it. Want the full analysis?Submit one yourself.
More from this channel
Jeff Dean: The 1% Rule for Building in AIY CombinatorJeff Dean links Google Search, TPUs, Gemini agents, and founder strategy to 1,000x energy gaps and 0%-success niches.
Patrick Collison: Is AI Breaking the Lean Startup Playbook?Y CombinatorPatrick Collison says Stripe took two years to launch, while Stripe data shows AI-era startups nearly doubled year over year.
Waymo Co-CEO Dmitri Dolgov: The Demo Is Only 1% Of The WorkY CombinatorWaymo’s demo took 18 months, its product 15 years, now reaching 17× human safety across 220 million autonomous miles.
Max Hodak: Average Is Not Good EnoughY CombinatorMax Hodak explains how Science’s Helix infrastructure, hiring system, and retinal implant work turn speed into startup advantage.
Peter Steinberger: "Fun Is Velocity"Y CombinatorOpenClaw grew from a WhatsApp relay into a 4.7-million-download open-source storm—and taught Peter Steinberger that fun drives velocity.
Related analyses
Rebuilding Git for AI Agents and The Future of Developer Tools | Deep Dives with a16za16z Deep DivesGitButler rethinks Git for AI agents with parallel branches, agent-ready CLIs, and patch-based code review
Datadog CISO on Securing AI Agents at Scale | Deep Dives with a16za16z Deep DivesDatadog expands AI from 50 Cursor licenses to 4,000 engineers, securing agents with MCP servers and an intent judge
Inferact: Building the Infrastructure That Runs Modern AIa16z Deep DivesInferact founders explain vLLM’s rise, 500,000-GPU scale, and a universal open-source inference layer
股市暴跌，無人消費：AI贏了，但白領消失了？！《全球智能危機》Better Leaf 好葉《2028年全球智能危機》警告AI裁員、13兆美元房貸與SaaS死亡螺旋將重塑經濟
AI Copilots Are a Dead End. Here's What Actually Works | Kavak CEOa16z Deep DivesKavak uses AI agents for 90–95% of customer interactions after a flat 2023, then grows four times.