Agnost AI: Catching Agent Failures Your Evals Miss in Real Conversations

Launch HN: Agnost AI (YC S26) – Extract user feedback from agent conversations

I built Agnost AI to solve a critical gap where your evaluations pass but production agents still fail. Our platform analyzes real user conversations to surface missed failures, extract feature requests, and even generate pull requests for fixes automatically. Backed by Y Combinator, we help teams like Google and Exa turn hidden feedback into actionable improvements, ensuring your agents get smarter faster without manual review.

"Agnost AI showed me users were asking for features we don't even have. I had no idea. All that feedback was just sitting inside conversations we already had."