Intelligent Data Dictionary Agent (Profiling-Grounded AI)
Link to open source: https://github.com/Joshuapj123/intelligent-data-dictionary-agent.git
Intelligent Data Dictionary Agent is an AI-powered system designed to automatically generate business-ready documentation from enterprise database schemas. The platform connects to a PostgreSQL database, extracts metadata such as tables and columns, and performs deterministic statistical profiling including null percentage, uniqueness ratio, and distribution metrics. This structured context is then provided to a Large Language Model (Gemini) to generate grounded, business-friendly descriptions, usage suggestions, and potential PII flags.
The goal is to bridge the gap between technical schema structures and business understanding by transforming static database metadata into structured, contextual, and data-aware documentation. Unlike simple chatbot-based schema exploration tools, our approach combines deterministic profiling with AI interpretation to reduce hallucination and improve documentation reliability.
This build was uploaded as a hackathon project
