Multi Agent Conversational Search

A Multi Agent Conversational product search system that uses multiple AI agents to provide personalized search results. The system combines vector search, natural language processing, and multi-agent coordination to deliver relevant product recommendations.

System Architecture

Below is the high-level architecture of our multi-agent system:

Demo

Features

• 🤖 Multi-agent architecture using CrewAI
• 🔍 Vector search with AstraDB
• 🎯 Personalized search results
• 💬 Natural language query processing
• 🎨 Modern, responsive UI
• ⚡ Fast API backend

Tech Stack

• Backend:

  • FastAPI
  • Python 3.9+
  • CrewAI
  • LangChain
  • OpenAI Embeddings

• Database:

  • AstraDB (Vector Database)

• Frontend:

  • HTML/CSS/JavaScript
  • TailwindCSS
  • Responsive Design

Getting Started

Prerequisites

• Python 3.9 or higher
• AstraDB account
• OpenAI API key

Installation

1. Clone the repository:

   git clone https://github.com/yourusername/multi-agent-conversational-search.git
   cd multi-agent-conversational-search

2. Create and activate a virtual environment:

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

3. Install dependencies:

   pip install -r requirements.txt

4. Create a .env file:

   OPENAI_API_KEY=your_openai_api_key
   ASTRA_DB_TOKEN=your_astradb_token
   ASTRA_DB_ENDPOINT=your_astradb_endpoint

Running the Application

1. Start the server:

   python3 -m uvicorn src.main:app --reload --port 8002

2. Open your browser and navigate to:

   http://localhost:8002
   

Project Structure

├── src/
│   ├── agents/
│   │   ├── planner.py
│   │   ├── retrieval.py
│   │   ├── personalization.py
│   │   ├── ranking.py
│   │   └── response_generator.py
│   ├── config/
│   │   └── settings.py
│   ├── models/
│   │   └── schemas.py
│   └── main.py
├── ui/
│   ├── index.html
│   └── static/
│       ├── script.js
│       └── styles.css
├── requirements.txt
└── README.md

Setting Up the Database

1. Create an AstraDB account and database
2. Create a vector collection:

   from astrapy.db import AstraDB
   
   db = AstraDB(
       token="your_token",
       api_endpoint="your_endpoint"
   )
   
   # Create vector collection
   db.create_collection(
       collection_name="product_search",
       dimension=1536  # OpenAI embedding dimension
   )

Search Examples

The system supports various types of natural language queries:

• Price-based: "gaming laptop under $2000"
• Feature-based: "laptops with longest battery life"
• Use-case: "best laptop for graphic design"
• Comparison: "compare MacBook Pro M2 vs Dell XPS 15"
• Brand-specific: "ASUS ROG gaming laptops"
• Specific Requirements: "ultralight laptop under 3 pounds with good battery"

Contributing

We love your input! We want to make contributing to Multi Agent Conversational Search as easy and transparent as possible, whether it's:

• Reporting a bug
• Discussing the current state of the code
• Submitting a fix
• Proposing new features
• Becoming a maintainer

Ways to Contribute

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

Show Your Support

If you find this project useful, please consider giving it a ⭐️ on GitHub! It helps the project grow and attract more contributors.

You can also:

• Share the project with others
• Create blog posts about it
• Mention it in your presentations
• Follow updates and releases

Development Discussion

Have an idea or question? Start a Discussion or join our existing ones. We'd love to hear from you!

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

• CrewAI for multi-agent orchestration
• OpenAI for embeddings
• AstraDB for vector search capabilities
• FastAPI for the backend framework
• TailwindCSS for styling
• All our amazing contributors! 🌟