AI-odyssey-Dione-hackathon

AI-odyssey-Dione-hackathon

This project demonstrates an Agentic RAG (Retrieval-Augmented Generation) system. It showcases how to build intelligent agents capable of retrieving information and performing tasks using vector search and large language models.

Tools & Technologies

The notebook utilizes the following libraries and tools:

• smolagents: Framework for building code-writing agents.
• Qdrant: Vector database for efficient similarity search and retrieval.
• Chonkie: A library for text chunking and data preparation (TextChef, RecursiveChunker).
• Hugging Face Transformers: Uses models like HuggingFaceTB/SmolLM3-3B.
• LangChain: Integrations for Qdrant and OpenAI (langchain-qdrant, langchain_openai).
• Rich: For beautiful terminal formatting and markdown rendering.
• PyPDF: For processing PDF documents (if you need to use PDFs instead of md files)

Contents

• Smollgents_with_Qdrant.ipynb : main notebook/demo using Qdrant.
• resources.md : additional resources and links.

Usage

1. Open the notebook with Jupyter or Google colab.

   Note: If you have difficulty viewing the notebook on GitHub, you can open it directly in Google Colab:

2. (Optional) Create a virtual environment and install dependencies.
   • Windows (PowerShell):

     python -m venv .venv
     .\.venv\Scripts\Activate.ps1
     pip install -r requirements.txt

3. Run the notebooks interactively.

Notes

• For the workshop recording please refer to this Drive link.
• For questions please feel free to reach out on linkedin Mohamed Arbi Nsibi

Acknowledgements

A huge thanks to Mehdi Bani, GDGoC FST, and Dione Protocol for the invitation!