Should you hand over your role with an AI agent trained on how you decide?

Answered by Falk Gottlob · Reviewed · 1 min read

THE SHORT ANSWER

Yes, as a reference for your reasons, and no, as the decider. Marco Di Santo's November 17, 2026 post describes an agent built on videos of himself explaining his choices, which colleagues can ask how he would have handled a new case. His five-question exercise is worth doing whether or not you are leaving. But an agent that answers what the person who left would do freezes their reasoning on their last day and leaves the decision without an owner. In my forward deployed series the test of leaving is the keys line, a named person who can change the system without calling you, and 7 of the 24 engineers in Milos Mandic's survey said their systems stay with them after handover. Add three fields to each decision: who decides now, what would flip it, and when it gets re-decided.

Build the agent if you have the two weeks. It is a better record of your reasons than a document, and Di Santo found it explained his older decisions better than he could from memory. Then finish the handover the agent cannot do: put another person's name on each decision.

The argument, the forward deployed version of the same problem, and the three fields are in Marco Di Santo Left an Agent That Decides the Way He Would. The Handover Needs the Name of Who Decides Now.. The day-60 page it borrows from is in Month Two: The FDE Leaves, the Workflow Has to Stay.

SOURCES

THE LONG VERSION

RELATED ANSWERS