Sentry under 8 mins
If you've never used Sentry, this is the short version. In under eight minutes, we walk through what it does, what it captures, and where to go next.
Sentry is an application monitoring platform. It collects errors, logs, timings, and replays, and ties them together with tracing, so you can go from "something broke" to the line of code that broke it without opening six tabs.
Here's what we cover:
- How Sentry groups failures into a single prioritized issue, and how the stack trace, logs, and tags point you to the cause. In our example, every affected user signed up with Google.
- How Session Replay lets you watch what the user saw, with text masked and images blocked by default.
- How a trace follows a request from the click to the database, and how Sentry flags N+1 queries and slow queries for you.
- How to monitor AI agents, including full conversations, tool calls, cost, and tokens. One example: an assistant said a task was done after the tool behind it failed with a 401.
- How Seer, Sentry's AI debugger, reads your issues, traces, logs, and code to find a root cause. It can stop at a plan or open a pull request for you to review.
- How to use Sentry from your coding agent through the CLI and MCP.
On data: out of the box, Sentry scrubs things like passwords, tokens, and credit card numbers. You decide what else gets collected, and you can choose US or EU storage.
Not sure where to start? Pick one product flow that really matters, like checkout or signup, and monitor that first.
Try Sentry for free: https://sentry.io/welcome
Docs: https://docs.sentry.io