🤖 ASADI: Intelligent Administrative Assistant (RAG)

An intelligent chatbot powered by Django and RAG (Retrieval-Augmented Generation) designed to query administrative and legal documents with precise context management.

📄 Project Documentation

For a deep dive into the architecture, the algorithms used, and the detailed results, please refer to our technical report:

👉 Read the Full Technical Report (PDF)

---

🚀 About The Project

ASADI is a full-stack application that leverages Large Language Models (LLMs) and ChromaDB (Vector Database) to assist users in navigating complex documentation.

Key Features:

• RAG Architecture: Retreives relevant context from indexed documents to ground LLM answers (reducing hallucinations).
• Workspace Filtering: Users can create specific workspaces to restrict the AI's search scope to a subset of documents.
• Dual Version Control: Synchronized environment bridging legacy SVN workflows with modern GitHub collaboration features.

👥 Team & Credits

This project was developed as part of the Computer Science curriculum at Université Paris Cité.

• Ali Traore
• Valentin Ponoussamy
• Thomas Gourmelen
• Abdel Malik Djaffer

---

🛠 Project Structure

The repository is organized as follows:

ASADI/
├── ASADI/          # Main Django Application logic
├── documents/      # Document ingestion pipeline & storage
├── documentation/  # Technical reports and project deliverables 📄
├── prompts/        # LLM Prompt Engineering & Templates
├── scenario/       # Pedagogical scenario generation logic
├── utilisateurs/   # User management & Authentication
└── workspace/      # Context filtering & Workspace logic