QueryBot: A Retrieval-Augmented Generation (RAG) System with Flask

QueryBot is an AI-powered chatbot built using a Retrieval-Augmented Generation (RAG) approach. It allows users to ask questions, retrieves relevant data from a database (Qdrant), and provides intelligent responses using OpenAI's GPT models. The system is designed with Flask, Docker, and an interactive user interface.

---

Features

• Retrieval-Augmented Generation: Combines retrieved data from a vector database with GPT for context-aware answers.
• Database Integration: Utilizes Qdrant, a high-performance vector database, for storing and retrieving data.
• Interactive User Interface: Simple and intuitive web-based chatbot interface.
• Context-Aware Responses: Tracks conversation history to handle follow-up questions effectively.
• Scalable Architecture: Deployed using Docker for seamless scalability.

---

Prerequisites

• Python 3.9+
• Docker (for Qdrant and deployment)
• OpenAI API key for GPT integration

---

Installation

1. Clone the repository:

   git clone https://github.com/ehsan5890/querybot.git
   cd querybot

2. Set up the virtual environment:

   python -m venv venv
   source venv/bin/activate  # On Windows, use venv\Scripts\activate

3. Install dependencies:

   pip install -r requirements.txt

4. Configure environment variables: Create a .env file in the root directory:

   OPENAI_API_KEY=your_openai_api_key
   QDRANT_HOST=localhost
   QDRANT_PORT=6333
   FLASK_APP=app/app.py
   FLASK_ENV=development
   

5. Run Qdrant (via Docker):

   docker run -d -p 6333:6333 qdrant/qdrant

---