To what degree is this a harness/system prompt problem? Models maybe should implement new stuff with as little impact on the existing stuff as possible by default? A simple system prompt for it to always check the code after task completion for proper simplifications, abstractions and cleanups before returning to the user? Instructions to retain "story like" readability of the code.
At least for Claude Code, putting "run /simplify at the end" in an "implement the plan" skill helps a little. It still often leaves new code in bizarre places, and/or with bad/alien-sounding names and comments.
I agree this is an option, and the next thing on my radar is to try with a more realistic "factory-shaped" harness where you have feedback from linters and other models after each coding episode that refines the architecture.
For readability specifically, I've found it hard to get the models to do this with prompting. If you've talked to opus/fable for a long time on prose writing you probably felt this too
A lot of the reason why concrete structures have failed so quickly is that the material was still fairly new when most of the structures failing now were created. There’s nothing that says that well designed and constructed modern high performance polymer modified concretes will have the same problems. They also have additives like zypex paste that make it more waterproof and self healing. They also have permeable formwork liners for increasing the surface density of the concrete and chemical concrete hardeners etc.
"A lot of the reason why concrete structures have failed so quickly is that the material was still fairly new when most of the structures failing now were created."
