CS undergrad at the University of Arkansas (B.S. expected May 2027), working on deep learning for 3D medical image segmentation.
Most of my time goes to brain metastasis segmentation — ensemble architectures over nnU-Net and SwinUNETR backbones, with a focus on small-lesion performance. I presented this work at the UAMS Radiology Research Symposium in April 2026.
Currently: running ANP, a pre-registered study of dynamic sparse training in 3D segmentation — testing whether a network reallocates capacity across layers better than nnU-Net auto-configures it. Writing up the metastasis segmentation results as a preprint.
Also: AlloScan (single-cell transcriptomics for renal allograft rejection), SwingScan (pose estimation), Nemesis (poker EV simulation).
Software engineering intern at Walmart Global Tech.

