LLM Policy

I generally try to avoid using generative AI (large language models; LLMs) in my personal projects. Sure, they can be useful for some tasks, but I find that I never learn much from using them, and I usually prefer to figure things out for myself. After all, the purpose of these personal projects is to learn and gain experience, and I can't do that if I'm relying on LLMs to 'think' for me like some delusional vibe coder.

Personally, I find LLMs good for some simple tasks or ideating, but often more trouble than they're worth for more complicated or novel problems. In an area that I'm familiar in, I find that it's a lot easier to do things myself than babysit an LLM through the process.

So, if some of these projects seem small or imperfect, that's because they were actually written by a human!