←back to thread

What makes Claude Code so damn good

(minusx.ai)
469 points samuelstros | 1 comments | 23 Aug 25 19:07 UTC | HN request time: 0.27s | source
Show context
athrowaway3z ◴[23 Aug 25 20:23 UTC] No.44998844[source]▶
> "THIS IS IMPORTANT" is still State of the Art

Had a similar problems until I saw the advice "Dont say what it shouldn't but focus on what it should".

i.e. make sure when it reaches for the 'thing', it has the alternative in context.

Haven't had those problems since then.

replies(1): >>44998950 #
amelius ◴[23 Aug 25 20:38 UTC] No.44998950[source]▶
I mean, if advice like this worked, then why wouldn't Anthropic let the LLM say it, for instance?
replies(1): >>45001801 #
donperignon ◴[24 Aug 25 06:09 UTC] No.45001801[source]▶
Because it’s embarrassing, and probably nobody understands why this works, depending on such heuristics that can completely change in the next model is really bad…
replies(1): >>45003104 #
1. amelius ◴[24 Aug 25 10:43 UTC] No.45003104[source]▶
I'd say exactly because behavior might change you have to include proper instructions for each model.

And depending on people in forums to provide these instructions is of course not great.