I think golang is likely a context rot issue. Sure its straightforward but the verbosity ends up affecting long horizon performance

Anecdotally Ive had decent output on a golang project but an incredible amount of duplication that I have to explicitly catch and have it correct.

What anecdotal experience? Not that I think LLMs produce good code, but for me it has produced a lot better c# than TypeScript fo example. I guess quality matters more than quantity when it comes to LLMs training data.