Building with Bots - Agent Monitoring in Sentry

Turns out people are building with AI a lot. It’s great, until you are the one that has to figure out why there’s a weird slowdown or a token spike after changing a model.

In this workshop, we’ll show you how to use Sentry Agent Monitoring to crack open the black box that is AI — inputs and outputs, token usage, model performance — so you actually know what the robots in your application are doing.

Key takeaways:

• Learn why AI monitoring is different and how Sentry helps
• Install the Sentry SDK and configure Logs for your AI app
• Enable Agent Monitoring and explore dashboards and spans to see exactly what your agents are doing
• Alert on slow model calls, rampant token usage, and tool call behavior