Work
Projects
Quantitative finance, machine learning, and applied mathematics — built end to end.
Work
Quantitative finance, machine learning, and applied mathematics — built end to end.
End-to-end ML pipeline predicting loan defaults with SHAP explainability and 0.788 AUC.
Automated ETL + Monte Carlo simulation mapping the efficient frontier across 5,000 portfolio allocations.
Normalized SQLite database with 10 analytical modules covering budget variance, cash flow, and anomaly detection.
Latent Dirichlet Allocation implemented from scratch via Gibbs Sampling with perplexity-optimized hyperparameter search.
Single and multivariable linear regression built from first principles using only NumPy.
Metaheuristic optimization applied to the NP-hard TSP — with direct parallels to portfolio rebalancing and execution optimization.