Comment by jwpapi
17 days ago
The first rule is an antipattern. I think describing your architecture or ANY kind of documentation for your AI is an anti-pattern and blows the context window leading to worse results, and actual more deviation.
The controlling systems are not give it more words at the start. Agentic coding needs to work in loop with dedicated context.
You need to think about how can i give as much intent as possible with as little words.
You can built a tremendous amount of custom lint rules ai never needs to read except they miss it.
Every pattern in your repo gets repeated, repo will always win over documentation and when your repo is good structured you don’t need to repeat this to AI
It’s like dev always has been, watch what has gone wrong and make sure the whole type or error can’t happen again.
it really does seem like this... also new devs are like that too: "i just copied this pattern use over here and there whats wrong?" is something i've heard over and over lol
i think languages that allow expression of "this is deprecated, use x instead" will be usefull for that too
I’ve written about that a bit here https://jw.hn/dark-software-fabric
AI gets a lot of big projects right if you give at all the tools to verify its own implementation if you can build a proper system to verify the solution it works astonishly good. Even Opus 4.6 judgement seems to be wrong most of the time on projects of my scale pre the validation layers.