Summary
Saviz Mowlavi is a research scientist in Cambridge, MA with eight years of experience developing hybrid model-based, physics-informed, and data-driven computational methods for faster prediction, optimization, and control of complex engineering systems. With a PhD from MIT and research stints at EPFL and MERL, he blends machine learning, model reduction, and numerical PDE techniques to tackle inverse problems, reduced-order control, and advection-dominated dynamics. His work ranges from analytical contact-force solutions for anisotropic particles to clustering algorithms that detect coherent motion in sparse, noisy physical data, showing strength across theory, simulation, and applied experimentation. He has repeatedly integrated deep learning and reinforcement learning with physical models for airflow estimation and control, and invented dimensionality-reduction and control schemes that outperform traditional approaches for waves and turbulent flows. Based in Cambridge, he brings a rare combination of mathematical rigor and practical engineering impact, often turning analytical insights into scalable computational tools. Notably, his research crosses domains—from granular avalanches to swallowing mechanics—revealing an ability to translate core methods to diverse physical systems.
8 years of coding experience
11 years of employment as a software developer
Doctor of Philosophy - PhD, Mechanical Engineering, 5/5, Doctor of Philosophy - PhD, Mechanical Engineering, 5/5 at Massachusetts Institute of Technology
Master of Science - MS, Mechanical Engineering, 5.8/6, Master of Science - MS, Mechanical Engineering, 5.8/6 at EPFL (École polytechnique fédérale de Lausanne)
English, French, Chinese