@yoheinakajima: get the most out of your agent traces w @agnostai
Summary
Agnost AI launched its first model, agnost-*******-0.1, trained on production traces from a customer, achieving a 22.9% increase in task success and a 90.2% reduction in some metric.
View Cached Full Text
Cached at: 08/20/26, 09:06 PM
get the most out of your agent traces w @agnostai
shubham (@ShubhamInTech): Stop using your agent logs just for debugging. Use them to train your own model.
Today we’re launching Agnost AI (YC S26)’s first model: agnost-*******-0.1
Trained for our first customer on their existing production traces & it beat the frontier:
+22.9% task success −90.2%
Similar Articles
@yoheinakajima: oooooh fancy new site? @shubhamintech
Agnost AI, a Y Combinator startup, has launched its first AI model, agnost-*******-0.1, trained on customer production traces, claiming significant improvements in task success rate.
@ShubhamInTech: Stop using your agent logs just for debugging. Use them to train your own model. Today we're launching Agnost AI (YC S2…
Agnost AI launches its first model, agnost-*******-0.1, trained on customer production traces to outperform frontier models with major improvements in task success, latency, and cost, offering custom fine-tuning services.
@yoheinakajima: if you build any agent on activegraph, the trace is automatic and first-class, not bolted on
Yohei Nakajima highlights that building an agent on activegraph automatically produces first-class traces, unlike bolted-on solutions, demonstrated with a coding agent experiment.
Agnost AI
Agnost AI is a product designed to catch failures in AI agent evaluations that traditional methods might miss.
@yoheinakajima: these guys built a research agent with activegraph (using their @monid_ai tool) and found that every claim was traced t…
A developer built a fully traceable and forkable research agent using Active Graph and monid_ai, ensuring every claim is natively traced to its source, and got it working in about 30 minutes.