Tracing
Log every step of a run: the prompt, each tool call with arguments and result, tokens, latency and cost. Tools like Langfuse or LangSmith show it as a clickable trace.
Last updated
Analogy
Browser devtools' network tab, but for your agent's thoughts.
In an agent
When someone says "the bot was wrong", the trace shows whether retrieval failed, the wrong tool was picked, or the model ignored the context.
Practice
Flashcards
- What is tracing?Recording every step of a run (prompts, tool calls, tokens, latency) for debugging.
Interview questions
- Your RAG gives wrong answers. How do you debug it?
- Check the trace: were the right chunks retrieved?
- If not, fix retrieval: chunking, hybrid search, re-ranking, metadata filters.
- If they were, fix the prompt or the model.
- Then add the failing case to the eval set so it stays fixed.
Questions people ask
What is tracing for AI agents?
Tracing records every step of an agent run: the prompt, each tool call with its arguments and result, tokens used, time taken and cost. Tools such as Langfuse or LangSmith show it as a clickable timeline, like the network tab in browser devtools, but for the agent's decisions.
How does tracing help when the agent gives a wrong answer?
The trace shows exactly where it went wrong. Maybe retrieval returned the wrong chunks, the model picked the wrong tool, a tool returned an error, or the model ignored the right context. Each cause has a different fix, and without a trace you would only be guessing.