This has been my experience. I tend to use chats, in a synchronous, single-threaded manner, as opposed to agents, in an asynchronous way. That’s because I think of the LLM as a “know-it-all smartass personal assistant”; not an “employee replacement.”

I just enjoy writing my own software. If I have a tool that will help me to lubricate the tight bits, I’ll use it.

Same. I hit Tab a lot because even though the system doesn't actually understand what it's doing, it's really good at following patterns. Takes off the mental load of checking syntax.

Occasionally of course it's way off, in which case I have to tell it to stfu ("snooze").

Also it's great at presenting someone else's knowledge, as it doesn't actually know facts - just what token should come after a sequence of others. The other day I just pasted an error message from a system that I wasn't familiar with and it explained in detail what the problem was and how to solve it - brilliant, just what I wanted.