IntelliSearchUsingOpenAI

A powerful Retrieval-Augmented Generation (RAG) search application built with Streamlit and OpenAI's Responses API. This application allows users to search through documents stored in an OpenAI vector store and receive AI-generated responses with proper citations.

🌟 Features

• Interactive search interface built with Streamlit
• Utilizes OpenAI's Responses API for intelligent search
• Supports multiple OpenAI models (GPT-4o, GPT-4o-mini, GPT-3.5-turbo)
• Configurable search parameters
• Displays search results with proper citations
• Maintains search history for easy reference
• Clean, responsive UI with custom styling

📋 Prerequisites

• Python 3.11 or higher
• OpenAI API key
• Vector store ID from OpenAI

🚀 Installation

1. Clone the repository:

   git clone https://github.com/mayankkapoor/IntelliSearchUsingOpenAI.git
   cd IntelliSearchUsingOpenAI
   

2. Create and activate a virtual environment:

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

3. Install the required dependencies:

   pip install -e .
   

   Or using the project's dependencies from pyproject.toml:

   pip install openai>=1.66.2 streamlit>=1.43.2 python-dotenv
   

4. Create a .env file in the root directory and add your OpenAI API key:

   OPENAI_API_KEY=your_openai_api_key_here
   

💻 Usage

1. Start the Streamlit application:

   streamlit run main.py
   

2. Open your web browser and navigate to the URL displayed in the terminal (typically http://localhost:8501)

3. Enter your search query in the search box

4. Optionally, adjust the advanced options:

   • Select the OpenAI model to use
   • Adjust the maximum number of results
   • Toggle whether to include search results

5. Click the "Search" button to perform the search

6. View the AI-generated response with citations and the original search results

🔧 Configuration

The application uses a vector store ID that is currently hardcoded in the main.py file. Modify this value to your own vector store ID.

🧩 Project Structure

• main.py: The main Streamlit application
• utils.py: Contains the RAGSearchClient class and utility functions
• static/styles.css: Custom CSS styling for the application
• .env: Environment variables file (contains your OpenAI API key)

📝 License

MIT License