Writing

Notes from the messy middle of AI and real work.

I write about agents, automations, and the awkward details that decide whether they survive outside a demo.

AI Operations · 7 min read

Jul 22, 2026

What to Measure After Shipping an AI Agent

After launch, the question is simple: did the agent save time, reduce mistakes, and earn enough trust to stay in the workflow?

Small teams do not need clever automations. They need workflows they can understand, debug, and safely turn off.

RAG works when you treat the document mess as part of the product, not as something the vector database will magically fix.

The agents I trust do not try to be heroes. They prepare the work, show their reasoning, and stop when the stakes get too high.

A blunt way to tell the difference between an agent that will earn its keep and a demo that will die after the first meeting.