Development now is often exploratory – making it work then making it right.
The verification steps are usually closer to QA than TDD.
Generating tests seem best for preventing regressions only after acceptance criteria is met.
I’m about to start banning agent’s from creating tests and see how it goes
Maybe the time saved from writing tests can be used to actually develop the product
Starting to think that CI is the new pre-commit hook for agents and needs to be seconds or less to run.
Until then, agents need to be notified so fixing CI isn’t polling or HITL.
