Writing / Topic

AI Engineering

Practical engineering notes on AI-assisted development, agent workflows, verification and security.

What this category covers

This category covers the engineering around coding agents: defining tasks, controlling permissions, preserving recoverable state, validating changes and deciding when a person must approve an action.

Treat proposed agent output as untrusted until it has been checked against the repository, tests and the relevant external system. Autonomy is not proof of correctness.

Articles in this category

1 article