transcribe

This AI monitors itself and opens its own PRs

LangChain · 0m · transcribed May 2026
More from LangChain
𝕏 Share ▶ YouTube 📥 PDF 🤖 .md

Transcript

0:00 One of our engineers, Alex Levenson, has created this like self-monitoring loop where he set up Inspect to run an action whenever there's a new PR that gets created or on the cron job, um, it will go and it will like look at the code base and try to see if like there's a monitor that is missing or there's like a metric that we should be adding. And then it goes into kind of like this this shadow mode where the monitor alerts won't like bug anyone yet, but then there's like another agent that will review them and if there's a they're like too noisy, then they get pruned because there's not enough signal. But if they run for a while and they're not noisy and the agent deems them to be like good enough, then they get like kind of promoted to start bugging the engineers. And also like kind of like slacks us the results.

Summary

Alex Levenson developed a self-monitoring system that automatically evaluates the codebase for missing metrics or monitors whenever a new pull request (PR) is created or during scheduled checks. This system operates in a "shadow mode" where alerts are initially suppressed, allowing for a review process to determine their relevance before they start notifying engineers.

- Self-monitoring loop initiated by new PRs or scheduled checks.
- System identifies missing metrics or monitors in the codebase.
- Operates in "shadow mode" to suppress initial alerts.
- Alerts are reviewed by an agent to filter out noisy signals.
- Monitors that prove useful are promoted to notify engineers.
- Results are communicated via Slack to keep the team informed.
© transcribe · For agents Built with care and craft by Gokul Rajaram