I am a computational scientist and HPC software engineer focused on integrating physics based modeling, machine learning, and cheminformatics to accelerate discovery in biology, chemistry and materials science.
At Berkeley Lab (JBEI), I lead the development of BayFlux, a genome scale Bayesian modeling framework that combines MPI, CUDA, and PyTorch for large scale metabolic flux prediction. I also lead the lab’s active learning driven therapeutic drug design pipeline, developing multi task learning models for potency, solubility and toxicity prediction using bioinformatics tools.
My work bridges HPC systems engineering and AI-assisted molecular design, building reproducible, containerized workflows that scale across supercomputers and cloud platforms. With a background in quantum materials simulation (DFT, GW, BSE), I am passionate about advancing how HPC and AI enable faster, more predictive science from drug discovery to renewable materials and beyond.