Research spans data-centric modeling across manufacturing and life sciences: ML and computer vision for condition monitoring; digital twin/surrogate integration; and robust, data-efficient learning with transfer learning, imbalance handling, and physics-informed validation.