Programming Horror

1935 readers

1 users here now

Welcome to Programming Horror!

This is a place to share strange or terrible code you come across.

For more general memes about programming there's also Programmer Humor.

Looking for mods. If youre interested in moderating the community feel free to dm @[email protected]

Rules

Keep content in english
No advertisements (this includes both code in advertisements and advertisement in posts)
No generated code (a person has to have made it)

Credits

Icon base made by Lorc under CC BY 3.0 with modifications to add a gradient

founded 2 years ago

MODERATORS

[email protected]

254

mallocPlusAI (sh.itjust.works)

submitted 4 months ago by sjmarf to c/[email protected]

31 comments fedilink hide all child comments

you are viewing a single comment's thread
view the rest of the comments

[–] [email protected] 57 points 4 months ago (18 children)

From my experience with ChatGPT:

It will NEVER consistently give you only the value in the response. It will always eventually add in some introductory text like it’s talking to a human. No matter how many times I tried to get it to just give me back the answer alone, it never consistently did.
ChatGPT is terrible with numbers. It can’t count, do math, none of that. So asking it to do byte math is asking for a world of hurt.

If this isn’t joke code, that is scary.

[–] [email protected] 3 points 4 months ago* (last edited 4 months ago) (7 children)

I know a guy who was working on something like this, they just had the call to the model loop until the response met whatever criteria the code needed (e.g. one single number, a specifically formatted table, viable code, etc) or exit after a number of failed attempts. That seemed to work pretty well, it might mess up from time to time but it's unlikely to (with the right prompt) do so repeatedly when asked again.

[–] [email protected] 5 points 4 months ago (5 children)

I'm currently a guy working on something like this ! It's even simpler as you can have structured output on the chatgpt API. Basically you give it a JSON schema and it's guaranteed to respond with JSON that validates against that schema. Spent a couple weeks hacking at it and i'm positively impressed, I have had clean JSON 100% of the time, and the data extraction is pretty reliable too.

The tooling is actually reaching a sweet spot right now where it makes sense to integrate LLMs in production code (if the use case makes sense and you haven't just shoe-horned it in for the hype).

[–] 31337 0 points 4 months ago (1 children)

This works well too, and with many different models: https://github.com/guardrails-ai/guardrails

[–] [email protected] 2 points 4 months ago

That's fucking badass thanks for the pointer this might prove useful. In the structured output department i'm hearing great things about dotTxt's outlines which lets you constrain output according to a regex, but i haven't tested it yet.

load more comments (3 replies)

load more comments (4 replies)

load more comments (14 replies)