Firefighting: The Engineer's Guide to On-Call
summary
100+ interviews across sizes and industries reveal concrete numbers and fixes: teams see 3 incidents/week on average and sometimes 200+ notifications/day, with 20% relying on useless human monitoring, 40% admitting muted alert channels and 70% depending on a single company hero. Common remedies are distributed on‑call across squads, a central reliability team, periodic tech‑debt sprints, a strict alert‑cleanup policy and internal automation; practical demos include auto‑remediation playbooks and AI agents that query DataDog, Grafana, GitHub and Slack via LangChain.
AI-written summary · about
chapters (7)
- 0:00Firefighting in engineering
- 2:28On-call as developer's life
- 6:04Heavy burden of on-call
- 8:56Incident response pain points
- 12:43Best practices for on-call stress
- 18:08AI agents in incident response
- 22:28Conclusion
related talks
24mWait for Us! Evolving On-Call as Your Company Grows
19mIts Time We Rethink On-Call Culture
06mIgnite: Key Factors for Successful 24×7 On-Call Duty
26mBuild a data-driven on-call workflow for your team with atomic habits
44mYou build it, you run it
29mDraining the Flood—A Combat against Alert Fatigue
31mHow to Build a Healthy On-call Culture
50mYou build it, you run it (why developers should also be on call)