PhD Candidate in Computer Science specializing in Machine Learning & Reinforcement Learning.
I'm a PhD candidate in Computer Science at the University of Dayton, researching reinforcement learning, large language models, and applied machine learning. Most of my work comes back to one question: how do learning systems behave under noise and uncertainty, and how do you build ones that stay reliable when the signal is messy?
Right now I'm working as a data science intern, building out analysis for the finance team alongside my research. Reasoning under uncertainty, filling out missing data, and disciplined empirical evaluation is what I do.
Outside the lab, I'm active in the Society of Women Engineers and volunteer with a 4Paws service-dogs training program!
Entropy-driven feature selection for adversarial robustness; 94–95% accuracy across Gym (LunarLander, BipedalWalker). Outperforms KL-divergence and joint-entropy baselines at detecting adversarial noise.
Modular glass-box vision-language pipeline for automated construction-safety analysis. Public summary only.
Classical search (DFS, BFS, UCS, A*) and RL policies (value/policy iteration, Q-learning) with trajectory visualizations.
Deep learning for time-series prediction. Preprocessing, tuning, LR scheduling, early stopping. Benchmarked vs MLP, CNN, and CNN-LSTM.
PPO agent that allocates a multi-asset ETF portfolio using a differential-Sharpe reward, benchmarked out-of-sample against equal-weight, mean-variance, and buy-and-hold with leakage-free backtesting.
Gradient-boosted default prediction on 30k accounts with imbalance-aware metrics, probability calibration, and SHAP explanations for auditable, per-applicant credit decisions.
Always happy to chat about machine learning, reinforcement learning, and applied data science. Résumé available on request.