alt Hacker NewsLLMs produce pretty crappy code but they are very useful tools for protyping, code search and finding bugs. Maybe LLMs in the future will be able to write good code but they are very far from that right now.
LLMs produce pretty crappy code but they are very useful tools for protyping, code search and finding bugs. Maybe LLMs in the future will be able to write good code but they are very far from that right now.
Perhaps it would be useful if both of you could provide examples of supposedly good and bad code – the latter being the result of a genuine effort to produce good code with state of the art models. Just asserting that LLM code is good or bad ends in a yes - no - yes - no back and forth circle immediately.