LightconePod AI agent infrastructure
Over the past year, we've been building our own internal agent infrastructure at YC: over 350 tools, self-improving skill loops, and a shared organizational brain that gets smarter overnight. In this episode of the @LightconePod, we sat down with YC General Partner Pete @koomen to talk about how he led the effort from the ground up. We cover how giving agents unrestricted access to one database was the key unlock, the self-improving skill loops that get smarter overnight, and why he thinks we've arrived at the personal computer moment for AI. 00:39 โ YC's AI Stack 02:15 โ The Finance Team Problem That Started It All 05:07 โ SQL Access Changes Everything 07:20 โ One Database to Rule Them All 09:14 โ Jevons Paradox 10:07 โ Denormalizing for Agents 12:15 โ The Single-Player Era of Agents 14:16 โ 350 Tools and a Shared Registry 16:24 โ Skillify, DRY, and MECE Resolvers 18:23 โ The Self-Improving Dream Cycle 20:26 โ The Two-Sentence Pitch Skill 23:06 โ How Super Intelligence Compounds 25:10 โ Recording Everything as a Building Layer 27:10 โ The Shared Organizational Brain 29:18 โ Trust-Default Culture as a Requirement 30:44 โ Raising the Floor for New Employees 32:35 โ Horseless Carriages 34:24 โ Why Chat Is the Best Interface for Agents 38:50 โ Just-in-Time Software 40:49 โ Centralizing vs. Decentralizing AI 43:32 โ The Personal AI Revolution