Simple RAG System (Chroma + LangChain + Google Generative AI)

This is a beginner-friendly Retrieval-Augmented Generation (RAG) system built using:
ChromaDB – Vector Database
LangChain – Retrieval + LLM Pipeline
Google Generative AI API – LLM for generating answers

The project demonstrates how to store document embeddings, retrieve relevant chunks, and generate accurate answers using RAG.

---

Features

🔹 Add your own documents for embedding
🔹 Store embeddings in ChromaDB
🔹 Retrieve most relevant chunks using similarity search
🔹 Use Google Generative AI to generate final answers
🔹 Simple, clean code for beginners
🔹 Easy to customize and extend

---

How It Works (Simple Explanation)

1. Load documents
   
2. Split into chunks
   
3. Convert chunks → embeddings
   
4. Store embeddings in ChromaDB
   
5. User asks a question
   
6. System retrieves the most similar chunks
   
7. Google LLM generates answer using retrieved text
   

RAG = Retrieval + Generation