Facio Blog

Practical notes on human-reviewed AI agents.

Payload-powered product notes, security writing, HITL patterns, and operational guidance from the Facio runtime: long sessions, Placet approvals, audit trails, memory, providers, channels, tools, and Docker-first operations.

Clear

Showing 21-25 of 57 articles in Human-in-the-loop.

Human-in-the-loop

The HITL Rollback Pattern: When Human Review Should Reverse an Agent Action

Jul 2, 2026Human-in-the-loop

The HITL Rollback Pattern: When Human Review Should Reverse an Agent Action

Most HITL designs treat approval as a one-way gate. The action executes, the audit trail records, and that's the end. But real systems need the ability to roll back — when a reviewer realizes they approved the wrong action, when an incident reveals a pattern, when a customer requests reversal. Here's how to design HITL rollback as a first-class pattern.

Human-in-the-loop

HITL Drift: Why Your Approval Gate Becomes Theater After 90 Days in Production

Jul 1, 2026Human-in-the-loop

HITL Drift: Why Your Approval Gate Becomes Theater After 90 Days in Production

Every HITL system starts with rigor and ends with rubber stamps. The reviewer who approved carefully in week one is clicking through 200 approvals per day in month three. The policy that was careful in version 1.0 is full of exemptions in version 4.2. Here's how HITL systems drift into theater — and the early warning signs that catch the drift before it's too late.

Human-in-the-loop

HITL Is a Contract: The Seven Clauses Between the Agent, the Reviewer, and the System

Jun 29, 2026Human-in-the-loop

HITL Is a Contract: The Seven Clauses Between the Agent, the Reviewer, and the System

HITL is sold as a feature, but it's actually a contract between an AI agent, a human reviewer, and the system that connects them. When that contract breaks, every party loses. Here's the seven-clause contract every HITL deployment needs in writing — and what happens when each clause is missing.

Human-in-the-loop

HITL and Model Versioning: How Approval Patterns Change When the LLM Changes

Jun 28, 2026Human-in-the-loop

HITL and Model Versioning: How Approval Patterns Change When the LLM Changes

You upgraded the LLM from claude-opus-4-7 to the next generation. The HITL override rate spiked from 4% to 12% overnight. The same reviewer pool is rejecting 3× more actions. The same policy is firing 5× more synchronous reviews. Model upgrades don't just affect outputs — they affect the entire HITL calibration. Here's how to manage the version transition.