Raindrop's founders on the safety layer for AI agents

Key points

Key takeaways from a YC Root Access Founder Firesides interview with Raindrop founders Zubin Koticha and Ben Hylak (September ư2026):

Transcript unavailable — summary based on title and description only.

Safety layer for agents. Raindrop detects when AI agents go wrong in production — from failed tool calls and hallucinations to problems companies didn't even know to look for.

Dogfooding origins. The tool started as a way to debug the founders' own coding agent, then became the company.

More capable agents, more failures. Better AI agents actually create more ways for things to go wrong, which iswhy the monitoring category exists.

Simulations. Raindrop's new simulations product catches failures before they reach production, going beyond what evals miss.

Traction. Already used by companies including Vercel, Clay, Framer and Speak, and just raised a Series A bringing total funding to $50 million.

Read more: YC Root Access — Building the Safety Layer for AI Agents