Skip to content
Apply

What does human in the loop mean for AI agents?

Human in the loop means a person approves or reviews what an agent does whenever the work falls outside rules agreed in advance. We agree the rules in writing at kickoff, and anything outside them goes to a person first.

Human in the loop
A design where a person approves or reviews what an AI agent does whenever the work falls outside rules agreed in advance.

How we apply it

  • Rules in writing at kickoff: the one job, the limits, and what live means.
  • A supervised run before go-live: your team approves anything the agent is unsure about.
  • Approval gates for the actions that matter, such as booking a meeting or sending a reply.

Examples from our delivery record

  • A reservation agent for a hospitality group answers and sends on its own between 08:00 and 20:00 on one lane, while its wine lane drafts every reply for a person to approve.
  • A law firm's scheduling assistant books only after a person approves.
  • A law firm's inbox watch runs every 30 minutes and raises an alarm on payment-change requests.

Why it matters

An agent that makes judgment calls with nobody checking is not what we build. Speed comes from the agent doing the routine work, and trust comes from a person owning the exceptions.

  • What is an AI agent?An AI agent is software that does a job in steps, using your tools, inside rules a person agreed. A plain-English guide with real examples.
  • How we build an AI agent in 30 daysKickoff, build on your real cases, a supervised run, then live. How a Phase 1 agent goes from first call to production in 30 days.
  • AI agents, GDPR and the EU AI ActWhat we build and document so an AI agent fits GDPR and the EU AI Act: a written record of what it does, the data it touches and who checks it.
  • What is an Agent Enterprise?An Agent Enterprise is a full team of AI agents set up and run for a company, with a person accountable for each job and us tuning the runs.