Part 8 of 12 · 2 min read
Answering human-in-the-loop questions in an interview
How do you add human approval to an agent without blocking everything?
"Two separate decisions. First, what to gate: by consequence, so low-risk read-only tools run freely and irreversible or costly ones need approval. Second, how to wait: pause the run, save state durably, free the worker, and resume by run id when the decision comes. In the example's simulation a blocking worker took 1,663 minutes for 50 tasks at a 30-minute mean wait, against 272 with pausing." Follow-up: "What if the person never answers?" A time limit with a default of stop or re-plan, never approve; and on resume I re-check the premise, since 13 percent of approvals at that wait would be older than an hour.
Why not require approval for every action?
"Because the reviewer's attention is limited. In the example's model with attention halving at 20 requests per hour, gating all six actions per task caught 23 percent of harm and leaked 71 per 1,000 tasks, against 60 for gating only irreversible actions. That fatigue curve is my assumption, but the effect is well known: rubber-stamping.
How do you decide what needs approval?
"I classify each tool by reversibility and cost, and gate the irreversible and costly ones, optionally adding a rule on a risk score. The risk score from the agent itself is only useful if it is calibrated, so I test it against outcomes; the example's simulation's best policy depended on an assumed informative score." Follow-up: "How do you know the gate is right?" I sample auto-approved actions for hidden harm, track the approval and edit rates, and check missed risks in test cases.
How does the human role change as the agent matures?
"From executor to reviewer to governor: people first do the work, then review exceptions, then set policy and audit logs. In practice I lower the gate threshold gradually, with evidence from the metrics, and keep a sample of audits." Follow-up: "What stops that from creeping unsafe?" Thresholds change only through reviewed policy, with a log of escalations.