This project demonstrates a complete end-to-end data pipeline. I took raw wildlife census data, performed heavy data engineering in a MySQL environment, and built a custom, programmatic dashboard using Python.
- Database: MySQL (Data Cleaning & Aggregation)
- Language: Python (ETL Pipeline & Visualization)
- Libraries: Pandas, SQLAlchemy, Seaborn, Matplotlib
- Tooling: VS Code, Jupyter Notebooks
- Extraction: Ingested 3,000+ raw records into a relational database.
- Transformation (SQL): Handled sparse data (NULLs), standardized date strings, and converted text-based booleans into binary integers for mathematical processing.
- Visualization (Python): Established a live SQLAlchemy connection to bypass manual exports and rendered a custom UI using
GridSpecfor precise layout control.
- The "NY Attitude": 63% of squirrels completely ignored human presence.
- Peak Activity: Foraging and eating behaviors spike significantly during the PM shift.
- Top Real Estate: Hectare 14D is the most densely populated grid in Central Park.
