> ## Documentation Index
> Fetch the complete documentation index at: /docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Instrument an agent and review traces

> Instrument your application so it streams traces into Arize AX, then review the traces a live agent produces.

Instrumentation captures each step your application takes and streams it into Arize AX as traces. With a few lines of setup, every LLM call, tool invocation, and retrieval becomes visible and ready to evaluate.

## Set up LLM tracing in minutes with OpenTelemetry | Ep. 3

[Set Up LLM Tracing in Minutes with OpenTelemetry | Ep. 3](https://www.youtube.com/embed/GRuItd_7raY)

What tracing requires and how instrumentation captures each step of your application.

## Build a real AI agent (financial analyst) with the Claude Agent SDK | Ep. 4

[Build a Real AI Agent (Financial Analyst) with the Claude Agent SDK | Ep. 4](https://www.youtube.com/embed/Kx33ExJkfmc)

Instrument a working agent end to end and review its traces as they arrive in Arize AX.

## Up next

> [card] [Create and validate evaluators](/docs/ax/learn/courses/getting-started-with-ax/observe-and-evaluate.md)
>
> Define success criteria and build code and LLM-as-a-judge evaluators you can trust.