Hi
I need to use the chatgpt api to chat and find the answers from my documents .
i am willing to spend on it if guidance are porvided to integrate that into my application
regrds
Ghazi anwer
You are looking for a RAG implementation. The first questions you must ask is where do these documents live. Then, you can choose an architecture that conbines indexing, search (retrieval), and generative ai to use the results from your retrieval.
The easiest prototyping tool right now is the Assistants API, though it’s in Beta and has a few issues to iron out. A more stable stack would be a vector db (Weaviate/Pinecone/Qdrant) + OpenAI (for embeddings & LLM), but you can also go open source on all fronts.
At Integrait Solutions we build custom assistants for clients in the US and EMEA, drop us a note if you’re interested on something production-ready. If you want to play around there are many tutorials out there.
Best of luck!