I build machine learning models that turn messy single-cell genomic data into biomarkers that researchers and clinicians can trust by combining optimal transport theory, Bayesian inference, and graph neural networks to quantify uncertainty in cancer biology.
At Kansas State University, I develop computational frameworks that improve single-cell genomics analysis and surface disease-relevant biomarkers, along with the uncertainty behind them. My research combines optimal transport theory, Bayesian inference, active learning, and graph neural networks to solve real problems in cancer biology and genomics.
Before K-State, I built computer vision and embedded systems for sports biomechanics at IIT Madras, and spent two years writing 60+ technical articles that translated complex science and technology for non-expert audiences, reaching 23,000+ readers. That habit of making dense material legible now shapes how I communicate research across disciplines.
I'm completing my PhD in Biomedical, Electrical & Computer Engineering and looking for my next role by applying research-grade computational biology to real problems in biotech R&D, healthcare AI, or clinical data science, where models need to hold up outside the lab.
The methods, tools, and problem spaces behind my research — the same stack I'd bring to a computational biology, bioinformatics, or clinical data science team.
Core research and technical roles. Leadership, writing, and mentorship experience is condensed below.
Alongside research, I've written for technical audiences and led teams — skills that carry directly into grant writing, cross-functional collaboration, and communicating findings to non-specialists.
I've written for outlets covering career development, AI/ML, cloud computing, and emerging technology — distilling topics like IoT, blockchain, and PCB design into clear, accurate, beginner-friendly writing. It's the same instinct I bring to explaining a Bayesian model to a wet-lab collaborator.
If you're building AI/ML solutions in biotech, genomics, or medical imaging — or hiring for a role that bridges computational research with real clinical and lab problems — I'd love to talk.