Developing warm-started randomized HOSVD algorithms optimized for GPUs to efficiently conduct lossy compression of streaming tensor data.

Empirical study on VIX-based dynamic hedging during macro shocks. Achieved 47% volatility reduction, 2.6× Sharpe improvement, and max drawdown from –19% to –4%. Top project in Erdős Institute Fall 2025 Cohort.

Implemented high-performance Monte Carlo pricing algorithms for exotic options in C++ and CUDA, achieving over 10000× speedup versus CPU baselines.

Developed automated large-scale data scraping and studied covariance shrinkage techniques for mean–variance portfolio optimization and evaluated their out-of-sample risk–return performance.

Developed a reusable library of numerical analysis and Monte Carlo algorithms with emphasis on stability, convergence, and reproducibility.

Conducted undergraduate thesis research on intersection theory and Schubert calculus, culminating in a defended thesis presentation.

A comprehensive classification of irreducible cubic curves in complex projective 2-space, exploring Bézout's theorem, Cayley-Bacharach theorem, and the algebraic and topological properties of elliptic curves.