Trails: Automated insights from LLM agent runs

Automatically analyzes your agent traces and surfaces insights. Helps you understand what went wrong and lets you create issues and fixes right away.

Hey all, Justin here. I previously built Phind, the AI search engine for developers. One of the biggest problems we had there was figuring out what went wrong with bad searches. We had tons of searches per day, but less than 1% of users gave any explicit feedback. So we were either manually digging through searches or making general system improvements and hoping they helped. This problem gets harder with agents. Traces are longer and more complex. I'm building a tool that lets you analyze LLM outputs directly to help developers of LLM apps and agents understand where things are breaking and why. I've put together a demo using browser-use agent traces: https://trails-red.vercel.app/vi... It's early, but I have lots of ideas - live querying of past failures for currently-running agents, preference models to expand sparse signal data. Would love feedback on the demo. Also if you're building agents and have 10k+ traces per day that you're not looking at but would like to, I'd love to talk.

Agent traces are brutal to navigate. You get a 40-step chain and somewhere in the middle a tool call returned garbage, but you're scrubbing logs manually to find it. Automated insight extraction from traces is the right move... curious how Trails surfaces the signal vs just showing a cleaner flame graph.