RAG Knowledge Engine
A small RAG application that answers questions about uploaded PDFs or text files by first searching the documents for relevant passages, then grounding an OpenAI GPT-4o answer in that retrieved text.
Visit website (opens in a new tab)github.com
RAG Knowledge Engine is a retrieval-augmented generation project built to reduce hallucination by grounding an AI's answers in a user's own documents. A user uploads PDFs or text files, and when a question is asked, the system searches those documents first to find the relevant passage before generating an answer, rather than letting the model answer purely from its own memory.
It uses OpenAI's GPT-4o as the language model, LangChain for the retrieval and generation logic, and ChromaDB as the vector store that indexes the uploaded document text.