41.8781° N, 87.6298° W — based in Chicago, working from anywhere
Data Scientist and Master of Applied Data Science candidate at the University of Michigan, with a math teacher's instinct for making complicated concepts click. I build with Python, SQL, and streamlit, and spend my off hours mapping everywhere I've traveled.
Paris
Ireland
Mexico City
detailed itineraries below
Global Travel Tracker
3D interactive globe →
*#GLB240124*
Passport Stamp Log
view collection →
*#PPT220056*
Statistical Modeling
explore models →
*#APSTATS260930*
Northern Lights Predictor
live aurora prediction map →
*#ICR009142*
Medellín, Colombia
The Big House, UMich
Lincoln Park Market, Chicago
Acropolis of Athens
Zaanse Schans, Netherlands
About Me
Before I wrote code and performed statistical analyses for a living, I taught the concepts behind my work to teenagers. I spent four years as a high school math and AP Statistics educator, then went back to school myself, pairing an M.Ed. and B.A. in Mathematics with a Master of Applied Data Science at the University of Michigan (3.93 GPA). Today I work as a Data Scientist, building tools based on ideas I used to teach my students.
My toolkit runs from statistical modeling and machine learning (scikit-learn, XGBoost, PyTorch) to dashboards people actually want to open (Streamlit, Tableau, Power BI), plus a growing interest in retrieval-augmented generation and LLM tooling. I care about clean systems, honest documentation, and explanations that meet people where they are, which is a habit that started at a whiteboard.
Boarding pass — chapter 1
Aug 2022 — Jun 2026 · Von Steuben Metropolitan Science Center (CPS) · Chicago, IL
Four years teaching Algebra 1, Geometry, Algebra 2, and AP Statistics, plus coaching cross country.
Boarding pass — chapter 2
Jun 2026 — Present · RCM Alternatives · Chicago, IL
Building automated SQL and Python pipelines that cut manual contract classification time by over 90%, and shipping Streamlit dashboards that give leadership a clear read on the contract categories.
4
years teaching in the classroom
3.93
GPA, U-M data science
Python
+ SQL, daily drivers
1
carry-on
Academic Background
Marist University
Mathematics
Foundation in rigorous mathematics, analytics, and core disciplinary studies. Graduated cum laude.
DePaul University
Mathematics
Honed instructional intuition and methods for making complex topics click.
University of Michigan
Applied Data Science
Advanced computational data science, machine learning, and scalable systems.
Expertise & Stack
four years of ap statistics, rebuilt for the web
A few of the interactive tools I leaned on constantly in the classroom. Reshape a Normal curve or a Chi-Square curve with a drag of a slider, or use the built-in Z-score and probability calculator to find P(Z ≤ z) below, above, or between any two values instantly. Built with Streamlit and Plotly, embedded live below.
live from the globe
A 3D interactive globe tracking every trip, stop, and route: flights as solid arcs, trains dashed, buses dotted, ferries as waves. Spin it, click a marker, and it pulls up the story behind that stop. Scoll down to see trip analytics. Built with Streamlit, embedded live below.
Paris, France
Summer Olympics '24
Banff, Canada
Lake Louise
Kenya
Safari
field notes
Every stop on the globe above, grouped by continent. Africa and Asia are next on the calendar ~ South Africa, Zimbabwe, Botswana, Madagascar, and Thailand are up ahead.
22
countries visited, 5 continents
Map data failed to load — check your connection and refresh.
North & South America — 4
Europe — 15
Africa — 1
Oceania — 2
Asia — 0
— next up, upcoming global expansion
geomagnetic forecast
Developed a real-time space weather forecasting tool using Python, Streamlit, and Cartopy that predicts geomagnetic storm intensity and dynamically maps the auroral oval. Trained and evaluated linear regression, Lasso, and LSTM models on nearly 20 years of NASA OMNIWeb solar wind data (~14 million rows), engineering 76 time-lagged features to forecast the Kp index across multiple horizons. The resulting lightweight Lasso model runs end-to-end in 0.6 seconds and achieves a test MAE of 0.58. The interactive web application integrates live NOAA feeds to render location-specific visibility probabilities and asymmetric auroral boundaries driven by spherical trigonometry and solar orientation shifts.
Banff
Canada
Waiheke Island
New Zealand
Bergen Fjords
Norway
Bláskógabyggð Waterfall
Iceland
Stade de France
Summer Olympics '24
Megalochori
Santorini
Monte Carlo
Monaco
Kenya
Safari