The mystery of the slow API (finding production bottlenecks in real time)

A user reports that clicking a button takes 8 seconds. Your CPU usage is fine, your database metrics look normal, and your server logs just say “200 OK”. How do you find the bottleneck without wasting hours guessing or littering your codebase with print statements?

In this interactive, hands-on workshop, you will step into the shoes of an engineer investigating a slow multi-service application. Starting with a broken codebase running in Docker, you will learn how to:

1. Spin up a local observability stack using OpenTelemetry and Docker Compose.
2. Auto-instrument microservices without intrusive code refactoring.
3. Trace requests across HTTP boundaries to uncover hidden latency in Jaeger.
4. Add custom context and error attributes to pinpoint root causes fast.

By the end of this session, you will walk away with a functional local OpenTelemetry setup and a GitHub template repository you can adapt for your own projects.