My background sits at the intersection of engineering, applied statistics, and finance with a B.A.Sc. in Engineering Science from the University of Toronto (concentration in mathematics, statistics, and finance, minor in AI), now an M.S. in Quantitative Finance at Fordham's Gabelli School of Business. I'm most interested in quantitative research: building models that hold up when you backtest and stress them, not just when you first fit them. The projects below are where I try to prove that, written in Python and C++, with the statistics shown alongside the code rather than hidden behind it.
Statistical Arbitrage Engine
Pairs/basket trading engine in C++: Engle–Granger and Johansen cointegration testing for pair selection, Kalman-filtered dynamic hedge ratios, GARCH(1,1) volatility forecasting, and VaR-based position sizing, backtested on an event-driven simulator.
Portfolio Tail-Risk Engine
Copula-based VaR/CVaR engine in C++ comparing Gaussian, Student-t, and Clayton copulas over GARCH-filtered marginals. Monte Carlo and historical simulation, backtested against realized breaches with Kupiec and Christoffersen tests.
NBA Betting Market Efficiency Model
Logistic regression and gradient-boosting models predicting NBA game outcomes from leakage-safe rolling team-performance and injury-absence features, benchmarked against de-vigged sportsbook lines via Brier score, log-loss, and calibration. A backtest on a held-out final season, run through a custom C++ engine, found no configuration with a real, validated edge against closing or opening lines. Market efficiency held up under testing.
Stochastic Modeling & Asian Option Pricer
Monte Carlo pricer for path-dependent derivatives (Asian, barrier options) using GBM simulation, benchmarked against Black–Scholes closed-form solutions, with 95% confidence interval accuracy checks and pathwise Greek estimation.
Bond Index Portfolio Optimizer
Optimizer minimizing tracking error against the ICE Treasury Bond Index using momentum ranking, dynamic asset allocation, and key-rate duration matching, achieving 3.7% annualized tracking error within 2.3% of benchmark return.
Product Researcher, Finance & Corporate Services
Built an AI pipeline automating QA for multilingual financial documents, integrating parsing and semantic validation for unstructured fund documentation, reaching 96% accuracy and 99.6% recall across 1,500+ documents.
Associate Solutions Manager, PreSales
Developed Excel proof-of-concept solutions supporting $1M+ USD in ARR for a FinTech SaaS client; used Power BI and SQL to turn static reports into dynamic insights, contributing to a 65% win rate.
Machine Learning Research Intern
Trained XGBoost and Random Forest models to predict transplant patient survival; used Bayesian hyperparameter optimization to reach an AUC of 0.70, under Professor Dionne Aleman.
- C++
- Python
- SQL
- MATLAB
- R
- NumPy / Pandas
- scikit-learn
- SciPy
- CVXPY
- Eigen
- Monte Carlo Simulation
- Black–Scholes
- VaR / CVaR
- Options & Futures Pricing
- Greeks
- Git
- Excel / Power BI
- Jupyter
- MySQL / Postgres