> I find LLMs very useful when I need to learn something new, e.g. a new library or API, or a new language. It is much faster to just ask the LLM how to render text in OpenGL, or whatever, than read through tons of terrible documentation.
LLMs are better at reading terrible documentation than the average programmer. So in that sense, "obscure text reading and explain it better" there seems to be a clear value add.
> If the answer is "a lot", you might consider how to solve that problem without relying on LLMs!
Aren't there languages with a lot of boilerplate though?
> LLMs are better at reading terrible documentation than the average programmer.
LLMs don't go and read the terrible documentation for you when prompted. They reproduce the information posted by other people that struggled with said terrible documentation, if it was posted somewhere.
It's still better than a modern web search or struggling with the terrible documentation on your own - for introductory stuff.
For going into production you have to review the output, and reading code has always been harder than writing it...
This is is wildly incorrect. Documentation categorically isn't excluded from LLMs' training sets, and they are very well able to summarize that documentation when asked.
LLMs are better at reading terrible documentation than the average programmer. So in that sense, "obscure text reading and explain it better" there seems to be a clear value add.
> If the answer is "a lot", you might consider how to solve that problem without relying on LLMs!
Aren't there languages with a lot of boilerplate though?