Hi, I'm Matt
Computer Science and Economics at CU Boulder. Currently seeking professional opportunities in Data and Analytics, Web Development, Machine Learning and Quantitative Finance.
.jpeg)
About Me
I'm based in New York City where I was born and raised, and am currently finishing my undergraduate studies at the University of Colorado Boulder remotely and building projects in Data and Analytics, Web Development, Machine Learning and Quantitative Finance. In my free time I enjoy reading, playing golf, and exploring the city, I also love watching the NY Rangers and the Knicks as well as College Football.
My academic focus sits at the intersection of data science and economics. Through my Economics minor I've built a foundation in micro and macroeconomic theory, and I like pairing that with technical work to apply quantitative methods to real economic problems. Long term, I want to design and deploy predictive and forecasting models that turn empirical analysis into clear, useful insight for economic and business decisions.
Education
University of Colorado Boulder
B.A. Computer Science, Economics Minor
Expected December 2026 · 3.5 Major GPA
Xavier High School
New York, NY
2018–2022
Technical Skills
Languages
Frameworks
Data/ML
Tools
My Work
Factor Risk Dashboard
Streamlit-powered portfolio analytics platform combining Fama-French factor models with a custom NLP-driven sentiment factor built from live financial news. Features single and multi-portfolio analysis with CAPM/FF5+momentum regressions, risk KPIs (VaR, CVaR, Sharpe, Sortino), and a FinBERT-scored sentiment factor updated daily via GitHub Actions.
RL Bond Market Making
Compares reinforcement learning agents against an exact finite-difference solver for optimal bond market-making under the Avellaneda-Stoikov framework. Implemented in both TensorFlow and PyTorch with matching methodology across single-bond and multi-bond portfolio settings.
Consumer Credit Risk Analytics
Builds a default prediction pipeline on 2.2M Lending Club loans. Engineers risk features from raw application data, compares logistic regression against XGBoost, and surfaces which borrower characteristics actually drive default risk through an interactive dashboard.
Grid Energy Demand Forecasting
Forecasts electricity demand for the PJM Interconnection using spectral analysis to extract cyclical patterns. Tests whether explicitly decomposing demand periodicity via FFT improves MAPE over calendar-dummy baselines, combining Python ML with MATLAB signal processing.
NYC Transit Efficiency Analysis
Investigates whether weather drives NYC subway ridership or if calendar structure (weekday, holiday) explains the variance. Built on MTA turnstile data and NOAA weather records, with an interactive dashboard visualizing the finding that weather effects are largely noise.
FlickPick
Team-built movie discovery app with a Tinder-style swipe mechanic for group movie night decisions. Built with a full-stack architecture as part of CSCI 3308 Software Development.
SkiWare
Ski repair diagnosis tool using Google Cloud Vision API for computer vision-based damage assessment. Built as a team project exploring cloud ML services.