I think you missed my point. Yes I can work on a library with an LLM, but I can’t work on the LLM itself, which is the interface I’m corresponding with now, where previously it was the library. Previously, the thing I was interacting with was malleable in a highly well-defined way. Now it isn’t.

Yeah: in a world where I am coding, I can put in the effort to make the software work the way I think... to use an LLM, I largely have to think the way it wants to work / converse.