Imagine you're a senior developer working on an account-management service. You ask a coding agent to add audit logging when users change their account details. It has access to the repository, you give it a few constraints, and you get on with something else.

AI can now generate clean, well-tested code in minutes, but who ensures we're building the right thing? This article explores why engineering judgment, challenging assumptions, and understanding business requirements are becoming increasingly important as AI transforms software development..

The models will change. Your architecture shouldn't have to. Why engineering leaders should optimise AI systems for the cost of changing their mind, not for today's best model.