Over a two-month period, contributed to the dataforgoodfr/13_democratiser_sobriete and dataforgoodfr/13_reveler_inegalites_cinema repositories by building robust data and video processing pipelines. Developed a Python-based OpenAlex article scraper with modular metadata extraction, PDF download, and improved CI/CD reliability using GitHub Actions and YAML. For cinema trailer analysis, implemented an end-to-end video evaluation pipeline that integrates face detection, attribute classification, and confidence-aware metrics, enabling reproducible annotation and visual review of representation in media. Emphasized code quality through refactoring, input validation, and documentation, resulting in maintainable, reliable systems for data ingestion and computer vision-driven analysis workflows.