RAG-Powered Query Resolver with Google Gemini

Turn your static text files into a dynamic, conversational knowledge base. This project leverages the power of Retrieval-Augmented Generation (RAG) and Google's Gemini API to let you ask questions and get intelligent, context-aware answers directly from your own documents.

Overview

In a world filled with information, our most valuable data is often locked away in static files—research papers, meeting notes, and project documentation. Finding specific information requires tedious manual searches. This RAG-Powered Query Resolver is a powerful yet easy-to-use Python application that allows you to simply talk to your documents.

By combining a robust retrieval system with the advanced generative capabilities of the Google Gemini API, this tool creates a personalized query engine. It doesn't search the web; it searches your world, providing answers grounded in the content you provide.

Core Features

• Create Your Custom Knowledge Base: Simply add your own .txt files to the data/ folder. The application automatically processes them to build its understanding.
• Powered by Google Gemini: Leverages Google's state-of-the-art Gemini model for nuanced, human-like answer generation.
• Context-Grounded Answers: The RAG architecture ensures that all responses are based directly on the information present in your documents, dramatically reducing the risk of irrelevant or "hallucinated" answers.
• Simple and Accessible: Built with clear Python code, this application is straightforward to set up and run, making advanced AI technology accessible to developers and enthusiasts.

How It Works: The RAG Pipeline

This project uses a Retrieval-Augmented Generation (RAG) pipeline to provide accurate, context-aware answers.

1. Ingestion & Indexing: When you start the application, it reads all the .txt files in your data/ directory. It splits the content into manageable chunks and creates a searchable vector index—a smart map of your information.
2. Retrieval: When you ask a question, the system first searches this index to find the most relevant chunks of text from your documents. This is the "Retrieval" step.
3. Generation: The retrieved text chunks are then passed to the Google Gemini API along with your original question. The model uses this specific context to formulate a comprehensive and accurate answer.

Getting Started

Use this repository to get started:
"https://github.com/akanupam/RAG-based-Query-Resolver"