|
I enjoy applying my mathematics & data science knowledge to answer questions in diverse areas, especially across multimodal data and complex systems. My work uses machine learning, mathematical modeling, and statistical testing to study neuroscience, biology, the environment, and social systems. |
🧠 Murray Lab – Computational Neuroscience: Developed automated preprocessing pipelines to extract 15,000+ waveforms for multivariate analysis, and explored how genetic, physiological, and activity-based features explain differences in motoneuron recruitment. Currently developing evaluation pipeline and running simulations of PyTorch recurrent neural network models using the spinal electrophysiology waveforms aligned to locomotor behavior described above in this new (RNN repo).
⚡ FUSRP – Axonal Signal Propagation Modeling: Large-scale biophysical simulations of action potential propagation in geometrically heterogeneous axons using finite-difference solvers and reaction–diffusion PDE models to study how myelin geometry, nodal radius, and excitability shape conduction speed and failure (team repo).
🌱 Quack Hacks – Sustainability & Greenwashing Analytics: Hackathon project combining data analysis, NLP concepts, and interactive web visualization to surface sustainability metrics and potential greenwashing signals. Includes a public-facing Next.js dashboard (site) and companion frontend code (repo).
🌍 World Happiness Modeling: Statistical and machine learning analysis of global happiness indicators using ridge/lasso regression, ensemble methods, PCA, and clustering to uncover latent structure and key drivers across economic, social, and cultural variables.
🧬 WiDS Datathon 2025 – Multimodal Brain Modeling: Multimodal machine learning analysis of fMRI functional connectomes and socio-demographic metadata to predict ADHD diagnosis and sex, comparing gradient-boosted models, feed-forward neural networks, and graph neural networks to study how brain connectivity and demographic factors jointly inform neurodevelopmental diagnoses (team repo).
🌊 NSF REU – Kansas Biological Survey: Empirical dynamical modeling and Gaussian process–based analysis of long-term NOAA ecological time-series data, using cross-validated forecasting and perturbation-based simulations to characterize nonlinear population dynamics.
Programming Languages
Data Science, Visualization & Infrastructure
