|
|
|
|
|
|
by akasakahakada
1163 days ago
|
|
|
GPT-4 still suck at scientific programming where you have to invent every wheel from scratch. It will be fine if you consider generic unoptimized code productive. Maybe all we need is to finetune a LLM for code optimization. But first you need a large enough codebase handcrafted by phds. But again not many phd are even qualified for performant code. |
|
|