On-call with an AI agent in Slack: MTTR drops from 45 to 14 minutes
ParticularFan1122 · reddit · 2026-10-26
A dev on a 5-person on-call rotation for a B2B app shares how they now start every incident by asking CodeRabbit's Slack agent what changed. It pulls recent deploys, error spikes, and PRs touching the failing endpoint, and usually pinpoints the culprit; on confirmation it opens a revert PR under the author's GitHub account, going through normal checks.
Results: the last 6 incidents averaged 14 minutes from page to mitigation vs 45 minutes for the prior 6 — mostly eliminating the 20 minutes previously spent just finding the right PR.
It's been wrong once: it blamed a config change, but the real cause was a manually dropped database index, costing 10 minutes. The author now reviews the agent's reasoning before reverting, and asks where others draw the line between suggesting and acting in incidents.
Related event: Dev team cuts incident fix time from 45 to 14 minutes with Slack AI agent(2 posts)→
More from coding & agent
- Watch an AI agent build a Minecraft scene block-by-block in a 6-hour timelapse — flowersslop · 2026-10-26
- Kornia author: AI agents skip library docs and read code/docstrings instead — ducha_aiki · 2026-10-26
- Dev rewrites Spain's AutoFirma in Rust — works as a Claude/ChatGPT skill to sign PDFs — cuenca · 2026-10-26
- Dev drops Jira: AI made coding faster than ticketing, so he tracks work in PRs and docs — feregri_no · 2026-10-26
- Kubernetes works for vLLM inference, but agent sandboxes needed Modal's custom stack at 1M+ concurrency — NielsRogge · 2026-10-26
- badlogic on his 2026 agent workflow: "I tell them what to do, I yell when they do it badly" — steipete · 2026-10-26