A surprising application I use LLMs for now is creating prompts for it. It is absurdly effective and helps me validate that what I told it is what it heard. Maybe I am late to the party on this but its surprisingly effective.

I find myself confused about the 'vibe coding' arguments (both for and against). Its another tool in our toolbox. It has its strengths and weaknesses.

Rejecting it completely is folly and using it uncritically is naive...how is that different from any other new tech we have seen.