Steve Lattanzio
I’m an engineer by training and nature. I work at the intersection of machine learning, forecasting, and risk, currently as a Principal Machine Learning Engineer at MetaMetrics (psychometrics and edtech) and Chief Innovation Officer at Weaver Mine Capital Management (quantitative finance). If you need a buzzword: AI.
I spend a lot of time designing quantitative systems from first principles. I’m especially interested in prediction: where it works, where it breaks, and what those failures reveal about the underlying system.
This site is a working portfolio and thinking space—projects exploring uncertainty, truth-seeking, and how to present complex information with maximum clarity and minimal distortion. It's my form of doodling (though I am known to doodle too).
Education
Duke Football • Pratt Fellow
Selected Projects
The projects below are attempts to make uncertainty legible.
NCAA Analytics
A scientific framework for modeling game outcomes and deriving implicit team strength. Produces rankings, spreads and total scores for upcoming matchups, and market comparisons using live pricing data.
Weather Engine
Minimalist weather app designed to convey probabilistic information about the most important things in your local forecast.
Forecast →Truth Engine
A carefully-crafted LLM answering system that compresses information into dense, direct, and truthful answers with minimal distortion.
Query →