ScholarRAG
A hybrid Retrieval-Augmented Generation (RAG) system for research Q&A using FAISS-indexed academic papers with optional live Wikipedia retrieval.
Tech Stack: Python, LangChain, FAISS, OpenAI API, Wikipedia
Built a hybrid Retrieval-Augmented Generation (RAG) system that answers research queries using a FAISS-indexed vector store of academic papers with optional live Wikipedia retrieval.
- Implemented document ingestion, embedding, and context-assembly pipeline via LangChain with citation tracking.
- Exposed a REST API for research Q&A, achieving 99% retrieval accuracy on benchmark queries.
- Supported optional live Wikipedia retrieval to supplement indexed paper knowledge.