[wrs]

LLMs have limited working memory, like humans, and most of the practices that increase human programming effectiveness increase LLM effectiveness too. In fact more so, because LLMs are goldfish that retain no mental model between runs, so the docs had better be good, abstractions tight, and coding practices consistent such that code makes sense locally and globally.

[pmg101]

So are we basically saying that LLMs work most effectively on codebases that exhibit good quality coding practices, but are not themselves particularly good at creating such quality code themselves, since they were trained on all the code that exists.

I don't know what conclusion to draw from that. Maybe that there's no such thing as a free lunch, after all.

[wrs]

The conclusion I draw is that LLMs need a human expert with some taste and agency to empower/supervise them. Just like human dev teams do.

[safetytrick]

It's actually been useful for me to explain certain best practices now that I can show that the LLM cares.

Why is this name bad? Because an llm will get confused by it and di the wrong thing half the time.