I develop scalable machine learning systems to simplify complex simulations and data, improving efficiency and reducing costs.

I am a PhD candidate in Physics specializing in machine learning, simulation, and high-performance computing. My research models effect of protein-protein interaction on viral assembly and converts high-dimensional physical systems into efficient computational frameworks.

Recently, I developed a machine learning–optimized coarse-grained elastic network model for Hepatitis B virus assembly using OpenMM and NAMD. By learning parameters from all-atom simulations, I achieved:

• Reduced simulation cost by 94% (5,000 atoms → 300 beads)
• Achieved ~17× speedup while preserving key structural and dynamical properties
• Accelerated simulations by an additional ~20× via parallelization on HPC clusters
• Enabled ~50× better sampling of rare events (e.g., oligomer formation) using weighted ensemble methods

Beyond accelerating simulations, I also:

• Extended a C++ kinetic Monte Carlo framework to model drug-binding effects on viral assembly
• Built Markov State Models to extract interpretable assembly pathways
• Produced computational results that directly complemented and guided in vitro experiments

I design machine learning–driven models, optimize them for scalability, build supporting infrastructure, and extract actionable insights from large-scale computational systems.

My strengths include:

• Probabilistic & stochastic modeling
• Machine learning for structured scientific data
• Performance optimization & parallel computing
• End-to-end computational pipeline design

I excel in high-ownership, technically ambitious environments, especially in machine learning–driven startups where solving complex problems demands both mathematical rigor and practical execution.

I am seeking industry roles beginning March 2026 in machine learning, applied research, or computational engineering.