← Back to context

Comment by Animats

6 days ago

I tried Github Copilot, free mode, on #1. The Python code outsourced all the hard work to numpy and pymeshlab, which is fine. Copilot wrote code to generate triangles. It's a reasonable job. Copilot's knowledge of what to call is better than mine. I have to try it now, but it did OK.

On #2, it set up the problem, but bailed on the gap filling part with

   # (This is a complex step; see note at bottom.)

That's not surprising, since I didn't tell it how to solve the problem. Can any of the premium systems do #2?

5 comments

Animats

Reply
nickpsecurity  6 days ago

When I used it, Copilot was for doing autocomplete on little, routine procedures. Like a few operations done on a list. I used it to fill in the blanks of tedious stuff in what I was already coding myself. It worked well at that.

However, if it took creativity or real analysis, I had to throw the big models at it with a carefully-written prompt. You want to use their recent, best, big models. I used to include precise descriptions of data structures and function API's, too, which helped a lot. From there, tweak the description until you find pieces that get the result.

Another thing. It will eventually synthesize code that's close but not quite it. I found that asking it to make small changes, one at a time, would help for a while. Ex: "Modify that code to do X" or "...X in part Y." It eventually started hallucinating in loops. I always had to fix some amount of what it generated but it still saved time.

I hope that helps.

cap11235  6 days ago

https://gist.github.com/cpsquonk/e9a6134e78a2c832161ca973803...

I did Qwen3-256B (a free model, but you'd need a host for something that large, probably. I used Kagi) and Claude Code.

Curious how these look to you.

  • Animats  6 days ago

    It actually wrote out the code for all the hard stuff.

    I like the Python code which outsourced the hard stuff to existing libraries. The odds of that working are higher.

    Can you tell it to use the "glam" crate for the vectors, instead of writing out things like vector length the long way?

    (We now need standardized low-level types more than ever, so the LLMs will use them.)