@codinghorror One trick is to have one agent do the summarization, and another agent review the summary and try to find errors and omissions.
Conversation
Notices
-
Embed this notice
Evan Prodromou 🎏 (evan@cosocial.ca)'s status on Wednesday, 22-Jan-2025 03:44:26 JST Evan Prodromou 🎏
-
Embed this notice
Jeff Atwood (codinghorror@infosec.exchange)'s status on Wednesday, 22-Jan-2025 03:44:27 JST Jeff Atwood
the whole AI / LLM thing is really quite useful for summarization of long text / long discussions -- as long as you make sure it didn't hallucinate anything important. Depends on the text, I guess?
-
Embed this notice