It's not too obscure. It's also about the point where some coding LLMs get weak. Zig changes a lot. So LLMs reference outdated data, or no data at all, and res

It's not too obscure. It's also about the point where some coding LLMs get weak. Zig changes a lot. So LLMs reference outdated data, or no data at all, and resort to making a lot of 50% confidence guesses.