Am I alone in not prioritizing the quality of prose produced by my coding agent? My foremost and almost only concern is how well it can engineer software.

When you spend 8 hours a day reading it, it has a pretty big impact. At least to me, its style is exhausting. Also very important for software itself. Documentation, tickets, code comments etc

I agree it's useless for any final-draft user-facing copy. However, again, I'm much more concerned about a reliable software engineering process, which Claude (and me in the loop) gives me in a way I have learned not to trust (at least not yet) from others.

People love to complain, particularly if being anti-AI is a big part of their identity.