The advancement of machine learning (ML) applications necessitates the construction of end-to-end experimentation pipelines characterized by scalability, robustness, and reproducibility. This thesis aims to empower Data ...