Skip to content
View WillPaz16's full-sized avatar
💭
vibe coding throwing 95
💭
vibe coding throwing 95

Block or report WillPaz16

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
WillPaz16/README.md

Will Paz

Mathematics · Data Science · Baseball

mathematics + data science @ Miami University.

Exploring exciting and complex problems through mathematics, data, and machine learning

LinkedIn · ORCID · X · Resume


About Me

  • I'm a Master's in Mathematics student at Miami University, working on my thesis and graduating in December 2026.
  • Currently working as a Data Science Intern at Arhaus after spending two years as Director of Analytics with Miami Baseball.
  • My work tends to sit somewhere between the intersection of all of the quantitative methods spanning applied mathematics, statistics, machine learning, AI, and just about anything in between.

Professional Experience

Data Science Intern, Arhaus (2026)
Built an automated document-ingestion pipeline (PostgreSQL + React) for cross-team data workflows, deployed via Azure and Docker; used Snowflake for graph-based ecommerce basket analysis and NLP-driven customer sentiment modeling.

Director of Analytics & Head Student Manager, Miami University Baseball (2023–2026)
Led a team of 4 student analysts overseeing advanced scouting, pitcher development plans, and pitch design collaboration with coaching staff. Built and deployed R Shiny/Plotly applications for player analysis; developed Stuff+, Location+, and swing-decision models. See Miami University Baseball below for the tooling built during this role.


Tools & Skills

Python R SQL Git React FastAPI Shiny Plotly PostgreSQL Snowflake Docker Azure Claude

Statistical Modeling Machine Learning Gradient Boosting Clustering Dimensionality Reduction Optimization NLP & Embeddings LLMs AI Automation Knowledge Graphs Topological Data Analysis


Selected Work

Money Bull — Mathematical Sports Ranking
With Mia Adler, Ford McDill, and Tiffanie Ng: developed a novel linear-algebraic ranking model for professional rodeo, published in Mathematics and Sports and presented at JMM 2025. Paper

TDA Pitch Clustering — Topological Data Analysis
Applied the Mapper algorithm and clustering techniques to MLB pitch-level data to explore the structure and similarity of different pitch types. Live demo

  • Miami University Baseball — Built analytics tools, models, dashboards, and scouting resources using Trackman and other baseball data to support player evaluation and coaching decisions.
  • MoneyScore — With Harrison Cradduck, Joey Endres, and Sam Honicky: NHL performance-vs-pay modeling (MoneyScore composite metric, Elo ratings, XGBoost cap-hit prediction), presented at JMM 2026.
  • Spotify Taste Graph — Built a graph-based representation of music taste to explore relationships between artists, genres, and listening patterns.

Coursework Highlights

DatabasesCobalt Mineral DB: normalized MySQL schema, functional-dependency/normalization analysis, and stored procedures over USGS mineral-deposit data.

Machine LearningML Classifier Benchmark: compared 9 classification algorithms (XGBoost, Random Forest, ANN, and more) on windowed sensor data, with full hyperparameter tuning and cross-validation.

Applied MathSVD Image Compression and House of Reps TDA: linear algebra and topological methods applied to image compression and congressional voting data — the latter an early precursor to the TDA thesis work above.

OptimizationSchedule Optimization: with Joseph Follrath and Jackson Frey, a mixed-integer linear program (OR-Tools) that builds optimal four-year course schedules under prerequisite, corequisite, and credit-load constraints.

Pinned Loading

  1. moneybull moneybull Public

    Analysis of Colley, Massey, Keener, and PageRank ranking methods applied to PRCA bareback rodeo riders. Research from NSF-funded Summer@ICERM 2024, published in Mathematics and Sports Journal.

    Jupyter Notebook 1

  2. miami-baseball miami-baseball Public

    Trackman-based pitching/hitting analytics for Miami University Baseball: data cleaning, scouting reports, a Shiny dashboard, and exploratory Python analysis.

    R

  3. tda-pitch-clustering tda-pitch-clustering Public

    Applying TDA's Mapper algorithm to MLB pitch-level Statcast data, paired with a CatBoost Stuff+ model, to study pitch quality and outcomes.

    Jupyter Notebook

  4. spotify-taste-graph spotify-taste-graph Public

    Graph-based knowledge base of music taste — artists, genres, and listening history enriched via Wikidata and Last.fm to map influence and discovery.

    Python

  5. schedule-optimization schedule-optimization Public

    MILP-based four-year course scheduler with prerequisite/corequisite constraint modeling, built with OR-Tools. CSE 478 team project.

    Jupyter Notebook

  6. cradduhj/hockeyMoneyScore cradduhj/hockeyMoneyScore Public

    SLM 418 Final Project - NHL Performance Data

    Python