Why I still read the code
A sprint I considered finished after checking screens led to many hotfixes. I now read code to find paths to test, then check those paths and the actual UI separately.
#ai-workflow
Posts with this tag.
The development workflow I refined with coding agents
An agent reported a passing test that waited for text the screen never showed. Delegating implementation led me to read plans and diffs myself and turn recurring corrections into guidance for later work.
Making the next success take less effort
After ranking effortful experiences first, I wondered whether that should include work completed with little effort. My experience with coding agents gave me another way to read the question.
Keeping working rules from a prototype migration
A prototype migration involved asking an agent to build similar features repeatedly. I put recurring corrections into skills, then found that short, absolute rules could produce the wrong changes too.