AI / Machine Learning · Data Science — I build AI systems from start to finish: multi-agent document pipelines, fine-tuned and calibrated transformers, LLM evaluation, and behavioral prediction models.
Madison, WI · Open to relocation
I'm finishing an M.S. in Psychology: Data Science in Human Behavior at UW–Madison (expected December 2026), after a B.S. in Psychology with a neuroscience concentration at UW–Parkside. Before graduate school I spent five years in high-volume logistics and operations, working on quality assurance, documentation, and process improvement. That is why I care about error rates, reproducibility, and cost as much as accuracy.
I'm interested in AI systems that know when they are unsure: calibrated classifiers, evaluation you can trust, and pipelines that pass hard cases to a stronger model or a person instead of guessing. Outside of that, I like turning physics into interactive software, and I'm curious about how people think and act, which is what my psychology training is about.
Each project has a case study on the portfolio.
- Digital-Mailroom: my capstone with American Family Insurance, one LangGraph pipeline for document routing, prompt experiments, scoring, and deployment.
- Mailroom-ML: a calibrated ModernBERT classifier that handles routine documents and knows when to hand off to an LLM.
- Mailroom Reloaded: the mailroom rebuilt as one deployable service, with an audit log of every decision.
- RVL-CDIP-Classifier: zero-shot document classification with vision LLMs, with tracked prompt versions and cost per page.
- Mailroom-Corpus-EDA: the data engineering behind the mailroom dataset: 3,302 documents from public sources, with provenance, label separation and an audit pipeline.
- AeroGlobe: a 3D flight-tracking globe built around a static data pipeline; in active development.
- Psychometric Digital Twin: LLM "digital twin" personas against seven classical models on 241 survey respondents; the personas did not beat simple baselines.
- MORNINGSTAR: a courtroom of LLM personas that argue, vote, and rule on engineering decisions.
- Wave Laboratory: a real-time 3D ocean and ship-motion sandbox that runs in the browser.
- The Mandalay Bay: a satirical casino resort simulator, play chips only.
- Education: M.S. candidate, Psychology: Data Science in Human Behavior, UW–Madison (expected Dec 2026). B.S. Psychology, UW–Parkside (2020–2025).
- Now: building services at UW–Madison Recreation & Wellbeing and driving for Avis Budget Group in Madison, alongside my capstone.
- Research: undergraduate researcher in a cognitive neuroscience lab at UW–Parkside (2024–2025), working with EEG and presenting at the 2025 Midwestern Psychological Association conference.
- Before that: logistics coordinator at Meijer Distribution (2020–2025), promoted from Retail Loader.
This repository is the source for the site — see how it's built.