Rag

RAG-Based AI Assistant

A complete backend API built with FastAPI that implements a Retrieval-Augmented Generation (RAG) pipeline. This system ingests text documents, converts them into vector embeddings using Google's Gemini API, stores them in a Pinecone vector database, and allows users to ask context-aware questions based on the ingested data.

🛠️ Tech Stack

• Framework: Python, FastAPI, Uvicorn
• AI/LLM: Google Gemini API (gemini-embedding-001 for vectors, gemini-2.5-flash for text)
• Database: Pinecone
• Data Processing: NumPy, python-dotenv, UUID

---

🚀 Complete Setup & Installation Guide

1. Prerequisites

Before running the code, you need:

• Pinecone API Key: Create a free account at pinecone.io. Create an index named rag-agent with Dimensions: 768 and Metric: cosine.
• Google Gemini API Key: Get this from Google AI Studio.

2. Clone and Install Dependencies

Open your terminal and run the following commands to set up the project and install all required libraries:

# Clone your repository (or just navigate to your project folder)
# git clone <your-repo-link>
# cd <your-project-folder>

# Install all required Python packages
pip install fastapi uvicorn pydantic python-dotenv pinecone-client google-generativeai numpy