🧠 RAG-Based AI Assistant — Module 1 Project (AAIDC)
Overview

This project implements a Retrieval-Augmented Generation (RAG) based AI Assistant using LangChain, vector databases, and large language models.
The goal is to build a system that can load documents, embed them, store them in a vector store, and retrieve relevant information to answer user queries.

Objectives

Understand and implement RAG architecture

Use text chunking and embeddings

Store and retrieve data using ChromaDB

Build a simple AI assistant capable of contextual answers

Demonstrate a working end-to-end RAG pipeline

🏗️ System Architecture
User Query → Embedding → Vector DB (Search) → Relevant Documents
→ LLM Prompt + Retrieved Context → Final Answer

Key components:

LangChain for chaining LLM + retrieval

Sentence Transformers for embeddings

ChromaDB for vector storage

Python for application workflow

.env file for secure API key management

📂 Project Structure
├── data/
│ └── sample_docs/
├── src/
│ ├── vectordb.py
│ ├── rag_assistant.py
│ ├── app.py
│ ├── utils.py
├── .env.example
├── requirements.txt
└── README.md

Setup Instructions
1️⃣ Clone the Repository
git clone
cd rag-ai-assistant

2️⃣ Create Virtual Environment
python -m venv venv
venv\Scripts\activate

3️⃣ Install Dependencies
pip install -r requirements.txt

4️⃣ Configure Environment Variables

Create .env file based on .env.example:

OPENAI_API_KEY=your_key_here
GOOGLE_API_KEY=your_key_here
GROQ_API_KEY=your_key_here

📄 Loading and Embedding Documents

Example code snippet:

def load_documents():
folder = "./data"
documents = []
for filename in os.listdir(folder):
if filename.endswith(".txt"):
with open(os.path.join(folder, filename), "r", encoding="utf-8") as f:
documents.append(f.read())
return documents

Retrieval and Generation
results = assistant.query("What is this document talking about?")
print("Answer:", results)

The assistant searches the vector DB for relevant chunks and generates contextual answers.

📌 Key Features

Automatic document embedding

Chunking + metadata tracking

Vector similarity search

Multi-provider LLM support (OpenAI / Google / Groq)

Clean modular architecture

🚀 Results

The system successfully retrieves relevant context and generates improved responses compared to a non-RAG chatbot.

Examples:

User: “Explain Section 2 of the document.”
Assistant: (Responds with retrieved context + LLM output)

🔒 Security Practices

.env is kept out of GitHub (included in .gitignore)

.env.example is provided for safe reproducibility

No API keys are exposed in the repository

📘 Conclusion

This project demonstrates a fully functional RAG pipeline as required for AAIDC Module 1. It integrates modern LLM workflows with vector search to produce contextual, relevant answers.