Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I think one should just use GPT to generate the prompts so as to reduce the human input further still, a kind of gpt-gpt-prompt-engineer-engineer.


Isn’t that what this codebase is doing? I haven’t grokked it 100% yet.

Recently I’ve been trying to engineer a prompt that I intend to run 1k times.

Noticing GPT4 bug out on several responses, I’ve talked it through the problem more and asked it to rewrite the prompt. So an automated approach to help build better prompts based upon held out gold data is useful to me.


And a GPT to receive the output, write a summary, and insert it directly into a CEO's trash folder.


I think it's exactly what they do here


Is it? I thought it just ranked the (manual) prompts against each other.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: