Re:Anthropic: "Applicants should not use AI assistants"
by layer8 on Monday February 03, 2025 @07:40AM
(#42917595)
Testing usually isn’t enough if you don’t understand the solution in the first place. Testing is a sanity check for a solution that you do understand. Testing can’t prove correctness, it can only rind (some) errors.
LLMs are fine for inspiration in developing a solution.
I don't advocate blindly trusting LLMs. I don't either and of course test whatever it spits out.