Youssef Masmoudi

• Designed and implemented an end-to-end MLOps pipeline for logo detection, integrating DVC, MLflow, GitHub Actions, Docker, Great Expectations, and NannyML. • Ensured data versioning, experiment tracking, model traceability, and automated CI/CD workflows for reliable ML deployment. Implemented data quality validation and continuous model monitoring in production to detect data drift and performance degradation. • Developed and evaluated multiple logo detection approaches, from classical OpenCV-based computer vision methods to deep learning models (CNNs, YOLOv8/YOLOv9), significantly improving detection accuracy.

• Automated functional test suites with Robot Framework, enhancing testing efficiency. • Integrated testing into a Jenkins-based CI/CD pipeline, enabling continuous testing practices. • Containerized applications and testing environments using Docker for improved scalability. • Deployed solutions on Kubernetes to ensure reliable and repeatable test execution across environments.

• Designed and implemented a GitLab CI/CD pipeline for a testing platform, enhancing development efficiency. • Integrated SonarQube for automated code quality analysis, ensuring high standards in code quality. • Deployed and managed the platform using Docker and Kubernetes, improving system scalability and deployment flexibility.