The Peter O’Donnell Jr. Postdoctoral Research Fellowship Program provides funding for recent PhD graduates to perform high-level, computational research with exceptional faculty in an interdisciplinary environment.
This semester, five remarkable new postdoctoral researchers have joined the Oden Institute community for the 2022-2023 academic year.
William Sands joins us from Michigan State University where he completed his PhD in Computational Mathematics, Science, and Engineering with a thesis entitled, “Numerical Methods for the Evolution of Fields with Applications to Plasmas.”
He conducted research in the Department of Computational Mathematics, Science, and Engineering at MSU, developing massively parallel algorithms for solving PDEs using successive convolution, contributed to novel algorithms for high-order field solvers targeting plasma applications, and lead the development of a platform-agnostic software tool for field solvers.
At the Oden Institute, he will be focused on the development of scalable algorithms for radiation transport on heterogeneous computing platforms. This project is a vital component in tools currently being developed to simulate plasma torches. He will be supervised by Dr. Bob Moser and Dr. George Biros as part of the large collaborative effort in the Center for Predictive Engineering and Computational Science (PECOS).
Brajesh Narayan was conducting research at University College Dublin in Ireland. He completed his PhD in Computational Physics with a thesis entitled, “Long-time methods for MD simulations: Markov State Models and Milestoning applications to studies of KRAS and ABL proteins and their interactions.”
At the Oden Institute, he will be working on developing and applying advanced long-time methods for Molecular Dynamics, particularly the Milestoning Method. Working alongside Dr. Ron Elber, Narayan plans to design an algorithm to determine the optimal positioning of milestones that will minimize computational costs and statistical errors. He will then use this method to study and model the permeating mechanism of a novel anti cancer cell-penetrating peptide (NAF-1) through the plasma membrane of malignant cells with atomistic details.
Katherine J. Pearce
Katherine J. Pearce comes to UT Austin from North Carolina State University where she completed her PhD in Applied Mathematics with a thesis entitled, “Mathematical models and parameter subset selection techniques for fibrin matrix polymerization in a biomimetic wound healing system.”
She participated in a Global Sensitivity Analysis working group to design and implement novel algorithms for sensitivity analysis in mathematical modeling and has conducted research in network measure framework for topological data analysis at the Air Force Research Laboratory. Her postdoctoral fellowship will be a return to her alma mater of UT Austin, where she graduated in 2013 with a BS in Mathematics and a BA in English Literature.
At the Oden Institute, Pearce will primarily work on randomized algorithms for rank-structured matrix compression with Dr. Per-Gunnar Martinsson.
Ankit Chakraborty did his PhD in Computational Engineering Science at RWTH Aachen University in Germany with a thesis entitled, “Optimal approximation spaces for discontinuous Petrov-Galerkin schemes.”
He was a graduate research assistant at the Aachen Institute for Advanced Study in Computational Engineering Science at RWTH Aachen University, an organization that shares the Oden Institute’s interdisciplinary approach to furthering computational science and engineering.
At the Oden Institute, he will continue to conduct research on Higher Order Galerkin Schemes and Metric-Based Anisotropic Mesh Optimization.
Kevin Miller joins us from UCLA where he completed his PhD in Mathematics with a thesis entitled, “Active Learning and Uncertainty in Graph-Based Semi-Supervised Learning.”
He was a graduate research fellow in the UCLA Department of Mathematics and has industry experience managing a team of five data scientists conducting machine learning research at Owlet Baby Care, Inc.
At the Oden Institute, Miller will work alongside Dr. Rachel Ward in an investigation of sequential decision making in machine learning paradigms that effectively leverage geometric information about data - such as active learning, coreset construction and graph-based semi-supervised learning methods.