16 October 2026
Open-source tool released for tracing and debugging AI apps and agents
First reported
Deep Learning Weekly ran this on .
- A free, open-source tool now lets developers trace, evaluate and monitor AI applications.
- It covers three kinds of AI system: LLM apps (chatbots built on language models), RAG systems (AI that looks up documents before answering), and agents (AI that takes multi-step actions).
- The purpose is debugging: seeing step by step what an AI system did, and where it went wrong.
How it was covered
Deep Learning WeeklyEditorial team
Tracing, automated evaluations, and dashboards for LLM and agent applications