← All case studies
Y Y Combinator · ~250 staff · world's most famous accelerator

Their AI writes its own new tools. Overnight. While everyone sleeps.

YC built a tool registry that grows itself. Every time a partner asks for something the AI can't quite do, it logs the gap. Then overnight, it builds the missing tool — and ships it to everyone the next morning.

350+
tools in the registry
0 → 100
new tools while you sleep
24/7
self-improvement loop
The problem

Every great AI tool gets stale the day after you build it.

Garry Tan and the YC partners use AI to do everything: review applications, draft feedback, summarise office hours, surface pattern matches across cohorts. They built tools for each one.

And every week, a partner would hit a wall. "I wish I could ask the AI to do X." But X was a brand new ask — no tool for it yet. They'd write a one-off prompt, get a half-good answer, move on. The gap between what the AI could do and what they wanted it to do never closed. It just kept widening.

What they built

A registry that watches itself. And a "dream cycle" that closes the gap.

YC's system has two pieces. The first is a tool registry — every AI tool the partners use, stored in one place, versioned, observable. When a partner uses a tool, it logs what worked, what didn't, what the partner wished it could do.

The tool registry (Friday afternoon)
application-reviewused 47x this week ✓
draft-followup-emailused 23x this week ✓
summarise-office-hoursused 14x this week ✓
cohort-pattern-matchused 8x this week ✓
competitor-pulseused 3x this week ✓
⚠ gap:"diff this batch's pitch decks against last batch" (asked 3x, no tool)

The second piece is the dream cycle. Every night, the system reviews the gaps and tries to build the missing tool. It writes a draft skill, tests it against the requested examples, iterates, and proposes it for review.

Overnight, the dream cycle runs
22:00
Scan the registryFind all the "wished I could do X" gaps logged this week.
22:14
Draft a candidate toolFor each gap, write a prototype skill that would close it.
22:38
Test against the original asksRun the prototype on the examples partners asked about. Does the output look right?
03:12
Refine, iterate, retryIf it failed, rewrite. Sleep is cheap. Compute is cheap. Iteration is free.
07:00
Propose for reviewThe tool shows up in the partner Slack channel: "I built deck-diff overnight. Want to try it?"

The result: the registry grows itself. Monday morning, a partner shows up to find a brand new tool waiting. They try it. If it works, they use it. If it doesn't, the failure logs back into the registry and the next dream cycle takes another pass.

The result

350+ tools. The AI is smarter every day.

YC's tool registry passed 350 named, versioned, observed tools — almost all of them written by the dream cycle, not by hand. Partners reach for an AI tool first because one usually exists for whatever they need to do. And if it doesn't, they trust that one will tomorrow.

The gap between what the model can do and what we need it to do should close itself. — YC's internal AI principles
What you can steal

Treat your skill library like a living organism, not a static manual.

Most companies write a list of "AI use cases" once, then never update it. YC inverted that: their "AI use cases" are a registry that the AI itself keeps current. The gap between what the AI does and what the team wants becomes the input to building the next skill.

You don't need partner-level AI engineers to do this. You need a structured registry, a way to log gaps, and a nightly job that takes a swing at filling them. That's the kind of plumbing we build.