Hacker News new | past | comments | ask | show | jobs | submit login

Chunk the PDF text and create embeddings. Get cosine similarity between user query and each chunk, and send the top N chunks to OpenAI that fit within token memory.



This is the way.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

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

Search: