[lelanthran]

> Those that say it's made me more productive or that I no longer have to do the boring bits and can focus on the interesting parts of coding.

I wonder about that bit, TBH.

If you're 10x more productive at generating lines of code because you're mostly just reviewing, just how carefully are you reviewing? If you're taking the time to spec out stuff in great detail, then iterate on the many different issues with the LLM code, then finally reviewing when it passes the tests ... how are you getting to 10x and not 2x?

TBH, for those people who really are able to create 10x as much code with the LLM, their employment is actually more precarious than those who aren't doing that - it means your problem domain is so shallow that an LLM can hold both it and the code in a single context window.