When it comes to disputes about AI coding capabilities, I think a lot of people who either don’t follow AI capabilities developments or who aren’t particularly capable coders might be missing some context, so I wrote a post wherein I try to get an AI coding tool to reimplement a feature in my game then closely study the output to demonstrate how well it’s able to do that.
Very impressive.
I tend to avoid using LLM assistance, since nowadays I mostly program in a hobby capacity, but I’m especially impressed by how well it navigated a pretty obscure programming language (Odin) in what’s presumably a fairly messy codebase (No judgement towards you! I’ve just never seen a videogame with clean code, I think it just comes with the territory).
Based on what I’m seeing though, it seems fairly squarely cemented as a tool for people who already understand how to program for the near term future.
I try my best. It was interesting to see Claude grepping through my files, it’s impressive that it managed to locate the relevant camera code despite it being in the stage update function (though fortunately for it the update code was all in one place). I’m also impressed that Claude Code can write Odin well at all, I do use ChatGPT pretty often for miscellaneous self-contained scripts and whenever I try to get it to write Odin code it very often produces syntax mistakes all over the place (mainly because it seems to get confused between Odin and C syntax).
I’m surprised you came to that conclusion, though I can sort of see how just the main body of my post might have given that impression. As much as I wish that were the case, it does seem to me that tools like this are in no small part being sold as a way for rank beginners to produce functional products. And in my opinion, similar to many other domains, LLMs shine as a way for non-experts to produce something low-quality that nevertheless works at all (rather than as something experts use to enhance their productivity or the quality of their work).