Match each HITL pattern to its defining tradeoff

Attempt it

Pre-action approval

Post-action review

Interrupt on threshold

Quick reference

TL;DR

Human-in-the-loop patterns trade autonomy against safety. Pre-action approval blocks every step, post-action review trusts then audits, and interrupt on threshold pauses only when risk crosses a line.

Memory aid

Sign in for the mnemonic that makes this stick. See Pro pricing.

Easy to grasp

Imagine a new intern who can take real actions on your behalf. You have three ways to supervise them. The strictest is to make them ask permission before every single move, which is safe but painfully slow because they wait on you constantly. The loosest is to let them do everything and only read their report at the end of the day, which is fast but means a mistake is already done before you see it. The middle way is to tell them to act freely on small routine things but to stop and check with you whenever the action looks risky or they are unsure. Most real systems use that middle approach because it gives you speed on the easy ninety percent and a safety brake on the dangerous ten percent that actually matters.

Key concepts

Human-in-the-Loop Agent Safety Action Reversibility

Quick scan · 2–4 min

Everything important, quickly.

Frame the three patterns as a single autonomy versus safety curve, match each to its defining tradeoff, then explain how action reversibility and blast radius drive the choice, and close on calibrating the interrupt threshold and why production systems mix patterns per action class.

Side-by-side

Pattern	Autonomy	When a human acts	Best fit
Pre-action approval	Lowest	Before every gated action	Irreversible, high stakes
Post-action review	Highest	After the run, via logs	Reversible, high volume
Interrupt on threshold	Medium	Only when risk crosses a line	Mixed action classes

Real-world examples

Real products, models, and research that use this idea.

LangGraph ships an interrupt primitive that pauses a StateGraph before a chosen node, implementing pre-action approval as a first-class human in the loop checkpoint.
Claude's computer use deployments commonly gate irreversible desktop actions, such as confirming a purchase, behind a human approval step while letting navigation run freely.
GitHub Copilot coding agents apply post-action review by opening a pull request, letting the human review and merge diffs rather than committing to the main branch directly.
Fraud and payment platforms like Stripe Radar use interrupt on threshold, auto-clearing low-risk transactions and routing only high risk score ones to a manual review queue.

Interviewer follow-ups

What an interviewer would ask next. Try answering before peeking at the approach.

QHow would you decide which tools in an agent get pre-action approval and which run autonomously?

Sign in or go Pro to see the approach for this follow-up.

QWhat signal would you use to trigger the interrupt in an interrupt on threshold system, and how would you calibrate it?

Sign in or go Pro to see the approach for this follow-up.

QWhy does applying pre-action approval to every step often backfire on safety in practice?

Sign in or go Pro to see the approach for this follow-up.

60-second cheat sheet

The night-before-the-interview bullets. Scan these on the way to the call.

Sign in for 5 unlocks/day or go Pro for unlimited revision sheets. See Pro pricing.

References

Primary sources. Skim if you want the original framing.