Perhaps it would be useful if both of you could provide examples of supposedly good and bad code – the latter being the result of a genuine effort to produce good code with state of the art models. Just asserting that LLM code is good or bad ends in a yes - no - yes - no back and forth circle immediately.
I just used an LLM (along with my decades of operating system development experience) to create a macOS tool [0] that lets me see through windows, instead of having to continually command+tab between windows.
The solution required reverse engineering and internals knowledge that most human engineers don't even have.
The question is no longer "Can an LLM write code?". It can. The problem is that certain humans refuse to put in the effort required to properly utilize these tools.
[0] https://imgur.com/a/2CUEjmA