Free Lesson

Automating Evals With Claude Code + Phoenix

Part of The AI Evaluation Handbook

60 min

Feb 19, 2026 4:00 PM

Virtual (Zoom)

In this video

What you'll learn

Connect Claude Code to Phoenix observability data

Set up the toolchain to give Claude Code direct CLI access to your Phoenix traces, spans, and performance metrics.

Use CLI commands to fetch traces and debug agents

Learn important Phoenix CLI tools for retrieving observability data and identifying errors in agent-generated code.

Prompt AI to analyze system behavior in real-time

Practice natural language prompts that enable Claude Code to interpret traces, diagnose issues, and suggest fixes.

Why this topic matters

As AI agents like Claude Code write more of our code, they need the same visibility into system behavior that human developers have relied on. Observability data: traces, spans, latency and errors are no longer just for dashboards and human eyes. This hands-on session covers how to give Claude Code direct access to your Phoenix observability data via the CLI.