Projects
Macroeconomic Forecasting and Machine Learning Paper Replication
Research Assistant, Johns Hopkins University — Dr. Domenico Giannone’s Lab
An independent Python/PyTorch replication of a machine-learning framework for probabilistic macroeconomic forecasting — from FRED-MD data processing through linear and deep neural network quantile models, hyperparameter search, and recursive out-of-sample evaluation of unemployment forecasts.
Gradient Descent Convergence Visualizer
An interactive Shiny for Python app for exploring gradient descent on classic optimization test surfaces — adjust the objective function, learning rate, and starting point, and watch convergence (or divergence) happen in real time.
Try the live demo → · Read the full write-up →
Gale–Shapley Deferred Acceptance Algorithm Playground
An interactive Shiny for Python app for exploring stable matching in traditional Doctor/Hospital matching markets — build your own doctor/hospital matching market and watch the deferred acceptance algorithm run step-by-step with either side proposing.
Try the live demo → · Read the full write-up →
Exponentially-Fitted Simpson’s Rule for Present Value Estimation
Introduction to Computational Mathematics — Spring 2026
Developed and tested a structure-aware modification to Simpson’s rule for numerically integrating present-value cash flow models, achieving 10–20x accuracy improvements by incorporating the exponential discount factor directly into the quadrature basis.