News for hackers. Stuff that matters.
Re:Two things LLM coding agents are still bad at by ivape on Thursday October 09, 2025 @03:22AM (#45524574)
Well, you see it hallucinates on long precise strings, but if we ignore that, and focus on what it’s powerful at, we can do something powerful. In this case, by the time it gets to outputting the url, it already determined the correct intent or next action (print out a url). You use this intent to do a tool call to generate a url. Small aside, it’s ability to figure what and why is pure magic, for those still peddling the glorified autocomplete narrative.

You have to be able to see what this thing can actually do, as opposed to what it can’t.

Re:Two things LLM coding agents are still bad at by sebtron on Thursday October 09, 2025 @03:55AM (#45524810)
> Well, you see it hallucinates on long precise strings

But all code is "long precise strings".

Re:Two things LLM coding agents are still bad at by ogogmad on Thursday October 09, 2025 @06:47AM (#45525879)
He obviously means random unstructured strings, which code is usually not.
Re:Two things LLM coding agents are still bad at by grey-area on Thursday October 09, 2025 @12:08PM (#45529660)
I can’t even tell if you’re being sarcastic about a terrible tool or are hyping up LLMs as intelligent assistants and telling me we’re all holding it wrong.