The repository is named data-profiler.
The primary language of the repo is TypeScript.
Project start date was July 2026.
It is described as a powerful, enterprise-grade data profiling and visualization tool.
Live demo URL: https://data-profiler-puce.vercel.app/.
I chose a decoupled architecture to keep the UI responsive while the backend handles heavy data processing.
DuckDB was selected because it can crunch millions of rows in milliseconds, and I pair it with Pandas for convenience.
The frontend stack includes React 18, TypeScript, Tailwind CSS, Lucide React icons, Recharts and custom SVG, and AG Grid React.
The backend stack consists of FastAPI, Python 3.10+, DuckDB 0.10.0, Pandas and the Uvicorn ASGI server.
To start the backend locally I run: cd python pip install -r ../requirements.txt uvicorn server:app --port 8000 --reload
To start the frontend locally I run: npm install npm run dev