YC's Garry Tan: One person + 20 agents can outperform a big tech engineering department
garrytan · x · 2026-08-15
In an a16z interview, Y Combinator President and CEO Garry Tan shares key insights:
- The Palantir miss: He turned down an early role at Palantir, pitched personally by Peter Thiel with a $70k check, calling it a "$2-4 billion mistake" because he chased a Microsoft promotion instead.
- Chase what you know, not what's hot: He reflects on quitting web programming in 2003 before Web 2.0 to chase mobile, emphasizing first-principles conviction over following the crowd.
- Pure SaaS is losing its moat: Per-seat SaaS may not exist in 5-10 years; founders need data, network effects, or other moats. The old 10-20x forward-revenue rule no longer holds.
- A markdown file is an employee: With agentic coding, "any given person could be 400 of that person." He turns business processes into reusable skill files run on cron jobs; companies go "zero to $15M ARR in about four months with two or three people and a few hundred skill files."
- Token-maxing: Frontier chat products are cost-constrained, so serious founders should run agents like OpenClaw at full strength, loading 800k-1M tokens per request at $50-100k/year, to "live in 2028 today."
More from AGI Musings
- Anima Anandkumar: AI Models Lack Understanding of the Physical World — AnimaAnandkumar · 2026-08-15
- Multiagent Coordination May Require Solving Alignment Science Problems — sebkrier · 2026-08-15
- Former ARIA Director Davidad on Posthuman Futures: Wisdom and Cosmic Expansion — danfaggella · 2026-08-15
- Iron Man's Jarvis predicted LLM-to-AGI evolution and agentic coding — steipete · 2026-08-15
- Developer Complains About AI Slop, Uses Pangram to Block AI Content — AccBalanced · 2026-08-15
- OpenAI Reports GPT-5.6 Sol Score Triples on ARC-AGI-3 After Harness Change — daniel_mac8 · 2026-08-15