Human-in-the-Loop
Where the human sits in an agent workflow: the decisions that stay with a person, how the agent asks for them, and what happens to the answer afterward. Approval gates, question surfaces, and decision records.
1 post
- 13 min read
The HITL Playbook: How Governance Turned an AI Coding Agent into a 51-Minute Incident Response Team
A 500 error surfaced an 8-day silent WordPress compromise at 09:04. Containment at 09:55. Full sign-off at 12:30. Here's what AI-assisted incident response actually looks like in practice — the human-in-the-loop governance, the two-pass workflow, the moment the agent was confidently wrong, and the persistence vector I would have missed without it.