LLMs work best when the user defines their acceptance criteria first

Re:LLMs work best when the user defines their acceptance cri... by jshmrsn on Friday March 06, 2026 @09:18PM (#47283767)

Considering that a fleur-de-lis involves somewhat intricate curves, I think I'd be pretty happy with myself if I could get that task done in an hour.

Given a harness that allows the model to validate the result of its program visually, and given the models are capable of using this harness to self correct (which isn't yet consistently true), then you're in a situation where in that hour you are free to do some other work.

A dishwasher might take 3 hours to do for what a human could do in 30 minutes, but they're still very useful because the machine's labor is cheaper than human labor.

[ Parent | on HN ]

Re:LLMs work best when the user defines their acceptance cri... by marginalia_nu on Friday March 06, 2026 @09:22PM (#47283801)

I didn't provide any constraints on how to draw it.

TBH I would have just rendered a font glyph, or failing that, grabbed an image.

Drawing it with vector graphics programmatically is very hard, but a decent programmer would and should push back on that.

[ Parent | on HN ]

Re:LLMs work best when the user defines their acceptance cri... by zeroxfe on Friday March 06, 2026 @09:32PM (#47283853)

> TBH I would have just rendered a font glyph, or failing that, grabbed an image.

If an LLM did that, people would be all up in arms about it cheating. :-)

For all its flaws, we seem to hold LLMs up to an unreasonably high bar.

[ Parent | on HN ]

Re:LLMs work best when the user defines their acceptance cri... by marginalia_nu on Friday March 06, 2026 @09:46PM (#47283933)

That's the job description for a good programmer though. Question assumptions and requirements, and then find the simplest solution that does the job.

Just about anyone can eventually come up with a hideously convoluted HeraldicImageryEngineImplFactory<FleurDeLis>.

[ Parent | on HN ]