devn913/Simple_RAG — reverse-engineered prompt

Build me a simple web app where I can upload PDF files and then chat with them like I’m asking questions to the document.

It should have a clean page with a sidebar showing my uploaded PDFs, their processing status, and a main chat area. After I upload a PDF, the app should read it, break it into searchable chunks, store it locally, and let me ask questions. The answers should come from the PDF content as much as possible, using Google Gemini for the AI responses and embeddings.

Please make it work as a Django app with the usual setup commands, environment variable for my Gemini API key, and local storage for uploaded PDFs and vector indexes. Include basic error handling for things like missing API key, PDF still processing, failed processing, or hitting Gemini limits. Keep the UI modern but not complicated.

Look up current docs online if you need to, especially for Gemini model names and LangChain setup.