Medical Knowledge Chatbot (RAG-based): This project is a GenAI-powered chatbot designed to answer medical questions by leveraging Retrieval-Augmented Generation (RAG). It utilizes LangChain, OpenAI GPT, HuggingFace embeddings, and Pinecone vector database to process and retrieve information from PDF documents. The architecture involves embedding user queries, performing vector searches, retrieving relevant document chunks, and generating answers with citations using OpenAI GPT. The chatbot is accessible through a modern Streamlit interface, providing real-time Q&A with source references. Key Features RAG Pipeline: Combines retrieval and generation for accurate, context-aware answers. Source Attribution: Provides page-level citations for transparency. Fast Semantic Search: Utilizes Pinecone vector DB for rapid retrieval. Customizable: Easily adaptable with different models, chunk sizes, or additional documents. Modern UI: Interactive chat and index management via Streamlit. Tech Stack LangChain (RAG pipeline) OpenAI GPT (LLM) HuggingFace Transformers (Embeddings) Pinecone (Vector DB) Streamlit (UI) Python This project is ideal for showcasing in a portfolio as it demonstrates the integration of advanced AI technologies and a user-friendly interface to solve real-world problems in the medical domain.