I’m developing a tool for analyzing long PDFs full of tables, aiming for ~90% accuracy in answering queries from the data. My current approach is using RAG + Azure Document Intelligence, but I’m not sure it’ll match OpenAI’s built in Data Analysis custom GPT.
Anyone here familiar with building similar functionality or know of any available code/resources for this? Any insights would be appreciated! 
I did see an open source version of code interpreter a while back, but it sounds like you are building something larger than that, sounds interesting, I’m sure there will be others here building similar systems. jr.2509 for visibility.
Is this the one you’re referring to? GitHub - KillianLucas/open-interpreter: OpenAI's Code Interpreter in your terminal, running locally
Indeed, that looks like the one, yes.
(extra words for the bot)