AI SRE and Agent Handoffs: incident.io's Collaboration Model
summary
Incident.io’s AI SRE runs live investigations in Slack, pins a short summary and updates a threaded progress feed, while a separate human-facing summarizer (A/B tested) preserves readable bullets and citations; a desktop Notch UI plus cord launches a terminal skill that downloads an investigation sync tarball to a sandbox and keeps it synced. Alerts trigger heads-up posts only when a hazard message rubric passes (unknown to humans, relevant, confident, citable). Delegated fixes hand off full investigation data to agents like Cursor via MCP for high-fidelity PRs.
AI-written summary · about
chapters (6)
- 0:00AI SRE as a worked example
- 5:19Channels are the agent interface
- 7:24Designing summaries people actually use
- 9:42Progress messages with citations
- 15:15Be where engineers work
- 19:58Handoff to delegated coding agents
related talks
30mAI Agents for Incident Management
16mAI Agents for Incident Investigation: The Good, The Bad, and The Ugly
40mWhen Incidents Fix Themselves: AI SRE in action
34mAutomating SRE Operations with Multi-Agent AI
35mAgentic Incident Response
17mFighting AI with AI
42mAI for security - friend or foe?
26mAI & SRE: Embracing the Ultimate Frontier of Automation