A quantitative developer specializing in building high-performance models at the intersection of finance, machine learning, and network science. Passionate about leveraging Rust, Python, and GCNs to solve complex analytical problems.
University of Utah | Sep 2024 - Current
Conducted usability studies for human-computer interaction research and applied statistical methods to analyze behavioral data, improving experiment design and results interpretation.
Breath of Fresh Air AI Hackathon | June 2025
Achieved 0.872 R-squared accuracy in forecasting PM2.5 air quality by developing an award-winning GCN-LSTM deep learning model.
Lucid Software Programming Competition | Oct 2024
Secured first place out of 500+ participants, including Master's and PhD students from universities like Duke and UNC Chapel Hill.
International Contest in Modeling | May 2025
Developed a spatiotemporal model for Baltimore’s transportation network using Graph Neural Networks (GCNs), TCNs, and Neural ODEs to validate congestion predictions.
A Java-based machine learning library that provides a collection of algorithms for tasks like classification, clustering, and regression.
• Integrate powerful ML capabilities into your Java applications.
Developed a modular, event-driven back-testing engine in Rust with Python bindings for seamless interoperability.
• Implemented SMA crossover strategies, tick-level execution, and walk-forward evaluation.
Built an unsupervised learning pipeline for detecting market regime shifts using entropy features derived from SP500 correlation matrices.
• Applied PCA for dimensionality reduction and K-Means clustering.
Engineered a dynamic SP500 index tracker with historical GICS sector mapping, price normalization, and log-return alignment.
• Enabled index replication and custom benchmark construction.
Developed a spatiotemporal model for Baltimore's transportation network using GCNs, TCNs, and Neural ODEs.
• Validated congestion predictions and infrastructure impact simulations.