Writing
Notes on human-in-the-loop AI
Approval design, policy gates, and what it takes to let an agent move real money.
Automation bias is a design bug
Show an AI's recommendation the wrong way and your reviewers stop reviewing. Automation bias is not a training problem or a user problem. It is a layout decision.
The gate is a policy, not an if
There are two obvious ways to decide when a human is required, and both are traps: hard-code it and it rots; let the agent decide and it won't. The floor is the third way.
How did this come about?
A blog about the unglamorous half of AI engineering: what happens when an agent proposes an action that a person has to stand behind.
What belongs in an approval context payload
A reviewer who approves without understanding the decision is a liability. Five things every approval request should carry — and three it never should.
Why waitForEvent is not human-in-the-loop
Every workflow engine can pause. Pausing is the easy half. The hard half is everything between pause and resume — the half you end up building yourself.