• Co-led a team of students (from undergraduates to graduate researchers) in training hundreds of deep learning models for brain pathology segmentation, including tumors and stroke.
• Developed and submitted winning methods to international MICCAI challenges, achieving 1st place in the ISLES 2024 competition.
• Applied state-of-the-art segmentation algorithms to plant root imaging for biofuel research applications.

• Worked with the Theoretical Division on uncertainty quantification of equation-of-state models.
• Applied Markov Chain Monte Carlo (MCMC) and Bootstrap sampling techniques then trained a supervised machine learning models to classify physically feasible states.
• Gained experience with HPC resources and Linux-based workflows for large-scale computation.
• Presented results at a research symposium and published in Journal of Applied Physics.

• Designed mobile test bed system for transporting test articles in and out of thermal vacuum chamber.
• Used SOLIDWORKS for 3D modeling and design validation under extreme vacuum and temperature conditions
• During year-round internship, gained hands-on experience performing Computational Fluid Dynamics (CFD) simulations with Star-CCM+ for spacecraft thermal component design.
• Obtained DoD Secret Security Clearance

Developing deep learning models for use in medical imaging of the brain.

• Aim 1: Improving state-of-the-art model performance in low-resource settings. Projects include domain adaptation, transfer learning, and clinically guided preprocessing.
• Aim 2: Designing conditional generative models for medical imaging that leverage reasoning to produce physically feasible outputs. Ongoing project in collaboration with Microsoft Health.
• Aim 3: Creating surrogate models for real-time inversion of partial differential equations in Magnetic Resonance Elastography (MRE). This involves integration of Python, FORTRAN, and MATLAB workflows.

Working collaboratively in a multidisciplinary team consisting of neuroscientists, surgeons, engineers, and physicists.

Programming primarily in Python with PyTorch, and running large-scale experiments on high-performance computing (HPC) clusters in Linux environments.

GPA: 4.02/4.33 (summa cum laude)

• Thesis: Development of models for the velocity–pressure gradient correlations in incompressible wall-bounded planar turbulent flows using multiple linear regression
• Performed statistical data analysis of Direct Numerical Simulation (DNS) datasets from NASA’s Turbulence Modeling Resource. Applied regression methods to model turbulent flow phenomena relevant to aerospace applications.
• Developed interpretable and robust models validated across a wide range of flow conditions.
• Presented work at AIAA Aviation Conference 2022 and the 10th International Symposium on Turbulence, Heat and Mass Transfer.

Selected Courses:

• Deep Reinforcement Learning
• Geometric and Probabilistic Methods in Computer Science
• Computational Fluid Dynamics

GPA: 3.94/4.33 (summa cum laude)

Dual major with undergraduate thesis. Courses included:

• Turbulent Fluid Dynamics
• Partial Differential Equations
• Stochastic Processes with Applications