Do they go into the architecture and code structure, or more just the user-facing result? Coding agents do a lot of copy-paste or near equivalents and make way too much code to accomplish many things.
Sure, but when I tried to vibe code something in a language I didn't have experience with, and so didn't look at the code at all, I had to basically trash the codebase after a few hundred lines because nothing I'd say could make the LLM fix the problems, or if it fixed them, more problems would pop up elsewhere.
In my experience, if you can't review the code and point out the LLM's mistakes to it, the codebase gets brittle fast. Maybe other people are better vibe coders than me, but I never managed to solve that problem, not even with Opus 4.1.