Ph.D. Candidate
CurrentPh.D. candidate in the department of Mechanical Engineering and Applied Mechanics. Research focuses primarily on using machine learning in molecular dynamics methodologies to improve open boundary simulations and to apply these techniques to nanoscopic atmospheric phenomena.Research highlights include:• Developing complete fluids-based molecular dynamics simulation projects, including ideation and formation of methodology and mathematical basis, as well as, additional code sets, simulation runs, and data processing• Implementing machine learning code into C++ based open boundary molecular dynamics simulations software (LAMMPS) to produce computationally efficient simulations with reducedruntime (up to ~100x improvement when tested against certain comparable methods)• Constructing a supervised learning method using neural networks to predict atomistic forces and particle fluxes using Python and PyTorch• Gathering data by observing nanoscopic droplet water growth of aerosols in atmospheric conditions using atomistic simulations• Utilizing server-based and supercomputer-based operations to submit and manage 1000s of simulations and postprocess results
