The Agent that doesn't need a rewrite
The agent that survives its first year in production is usually a thin layer over the system you already have. Your ten-year-old codebase is not the obstacle. It is the asset.
The agent that survives its first year in production is usually a thin layer over the system you already have. Your ten-year-old codebase is not the obstacle. It is the asset.
Renaming tests to evals did not create a new discipline. It created permission to skip an old one - the obligation to define what correct means before shipping.
An agent is a production system that happens to be probabilistic, and it has to be operated like one - the surviving minority are not the teams with a better model, they are the ones operating better.
Every AI integration is a bet that you can monitor something you do not fully control.
As code generation gets cheaper, the real bottleneck shifts from production to coordination, comprehension, and judgment.
Newsletter
25 questions for teams running an AI agent in production — get it now as a PDF. You'll also get new Notes on Systems when something is worth sharing. No launch theatrics.
RSS is available too: subscribe via feed reader.