This reads like mentoring a junior dev. You do mentoring because the pay off is a junior that develops into a senior; writes better code more efficiently. But what's the pay off with going through this process with AI?<p>A few days ago I spent an hour trying to get Claude to write unit tests for a relatively simple function, without getting anything functional in return. It produced good (but changing) plans of work, but the code generated didn't match the plan (syntax or semantics).<p>Are people simply replacing 'time spent coding' with 'time spent mentoring AI'? And is that a useful trade off, given AI forgets everything?<p>I'd love to see a more honest discussion about the overhead of coding with AI assistance. It's still too high an overhead for me at the present capability.