معرفی
Philip Mocz is a computational research scientist at the Flatiron Institute's Center for Computational Astrophysics, part of the Simons Foundation. His work focuses on developing scalable, high-performance multiphysics simulation software with a growing emphasis on integrating modern AI techniques and automatic differentiability. Previously, he was a Computational Physicist at Lawrence Livermore National Laboratory, where he specialized in designing high-order Arbitrary Lagrangian-Eulerian (ALE) finite element methods for magnetohydrodynamics (MHD) simulations on heterogeneous computing architectures as part of the Multiphysics on Advanced Platforms Project (MAPP).
Dr. Mocz earned his Ph.D. in Astrophysics from Harvard University in 2017 under Lars Hernquist, where he developed a finite-volume moving mesh magnetohydrodynamics algorithm applied to study structure formation and magnetic field amplification, integrating his solvers into the Arepo simulation code. Prior to his doctorate, he received an A.B. in Mathematics and Astrophysics from Harvard in 2012.
His research spans multiphysics simulations, cosmology, galaxy formation, black hole physics, turbulence, numerical methods, and AI integration in computational astrophysics. A significant focus involves cosmological simulations of alternative dark matter candidates, particularly fuzzy dark matter. His work bridges theoretical astrophysics with high-performance computing, developing novel simulation frameworks that incorporate modern computational techniques.
Dr. Mocz's publication record reveals a strong trajectory in computational methods for astrophysical problems, with increasing integration of AI techniques in recent years. His research demonstrates expertise across multiple domains including quantum mechanics applications to cosmology, turbulence modeling, and the development of advanced simulation algorithms. The interdisciplinary nature of his work connects astrophysics with computer science and applied mathematics.
He maintains an active educational presence through his blog featuring approximately 100-line Python tutorials on scientific computing at the undergraduate level, published on Medium and followed on Twitter. His educational materials cover fundamental computational methods including finite difference approaches, Riemann solvers, and differentiable simulations using JAX.
Dr. Mocz has served as a Teaching Fellow for Harvard courses including Astronomy 151 (Astronomical Fluid Dynamics), Applied Computation 274 (Computational Fluid Dynamics), and Applied Mathematics 205 (Advanced Scientific Computing). His outreach activities include mentoring for the LLNL DSTI Research Program, NASA Cosmic Origins Transitional Leadership Team, and Princeton Astrophysics Undergraduate Summer Research Program.
Originally from Hawaii, he enjoys outdoor activities when not working. His professional presence includes active GitHub repositories (pmocz), a personal website (pmocz.github.io), and engagement on Bluesky (@philipmocz.bsky.social), where he shares insights about computational physics and scientific software development.
