In February 2026, George Ștefan contributed to the zahariesergiu/ubb-sociology-ml repository by building a Census data preprocessing pipeline to support machine learning workflows. He focused on transforming raw Census data into ready-to-use train and test datasets, enabling streamlined model training and evaluation. Using Python and leveraging skills in data analysis and preprocessing, he established a reproducible data engineering foundation that improves data quality and supports future ML development. His work emphasized structured version control, with clear commits tracking the addition of preprocessed assets. The depth of the contribution lies in creating reusable, maintainable inputs for predictive modeling and analysis.