News for hackers. Stuff that matters.
Re:6 Weeks of Claude Code by exe34 on Sunday August 03, 2025 @09:54AM (#44776602)
> can still be predicted and understood (no matter how difficult that is).

If we're making up hypotheticals, then LLMs can be predicted and understood (no matter how difficult that is). They don't run on pixie dust.

Re:6 Weeks of Claude Code by shortrounddev2 on Sunday August 03, 2025 @11:10AM (#44777092)
It's qualitatively different to go through source code and specifications to understand how something works than to look at a database with all the weights of an LLM and pretend like you could predict the output.
Re:6 Weeks of Claude Code by exe34 on Sunday August 03, 2025 @12:12PM (#44777544)
you don't have to - just run the matrix multiplication. it's no different from using a pocket calculator.