My RAG Project Demo

Retrieval-Augmented Generation (RAG) App Overview

This project implements a simple Retrieval-Augmented Generation (RAG) system using the following stack:

• LangChain for orchestration
• Pinecone for vector storage
• HuggingFace Transformers for embeddings
• Streamlit for the front-end interface
• Ollama for running a local LLaMA 3 model

It allows users to query a document collection (e.g., IRB policies and procedures) using natural language and receive context-aware responses from a local language model served through Ollama.

Folder Structure

rag_project/
├── .env                    # Environment variables (e.g., Pinecone keys)
├── app.py                  # Main Streamlit app that ties everything together
├── logger.py               # Custom logging logic
├── exceptions.py           # Custom exception handling
├── requirements.txt        # Python dependencies
├── README.md               # Project documentation
├── ingestion.py            # Logic for document loading and embedding
├── query.py                # Logic for answering user queries