At Pennsylvania State University, I study Markov chain mixing and phase transitions in spin models, combining theoretical analysis with empirical simulations. I also developed a Python algorithm for estimating Ollivier-Ricci curvature in hypergraphs; its closed-form lower bound achieved a median 450x speedup over optimal-transport solvers.
At Samsung SDS, I designed and implemented a secure electronic voting system using homomorphic encryption in C++, and developed cryptographic libraries for privacy-preserving machine learning and secure data analysis. My research also includes stable representations of topological and spectral information using persistent Laplacians.
