The amount of reference material it has is also a big influence. I’ve had to pick up PLC programming a while ago (codesys/structured text, which is kinda based on pascal). While chatgpt understands the syntax it has absolutely no clue about libraries and platform limitations so it keeps hallucinating those based on popular ones in other languages.

Still a great tool to have it fill out things like I/O mappings and the sorts. Just need to give it some examples to work with first.

There’s actually an HAI video on that. Those names were actually a direct result of an attempt by Amazon to curate their products better. https://youtu.be/_Bq-6GeRhys?si=ih1eyBLJwo7KAVuS

I definitely rely on documentation more than copilot, since I’ve noticed that the code it writes is only ever as good as your own codebase.

Most of the stuff I code is API wrappers to get arbitrary data into a format our broadcast graphics system can understand. Once all the data structures are properly defined copilot is extremely useful in populating all the API endpoints.

The actual problem solving is getting the data in the first place and morphing it into the correct format.

I’m very much a novice coder, but I often find myself doing the opposite. Write a good comment and let copilot write the actual code.