About

Who Am I?

Hi I'm Xavier Genelin. I'm a data scientist with 5+ years experience working with data and 3 of those being in the field of Data Science and Machine Learning. I have an educational background in mathematics, economics, and statistics. I received my bachelors in Mathematics from Xavier University and my masters in Statistics from North Carolina State University.

I have experience working with python, R, and SQL to solve problems and assist in the decision making process using data. I love the different problems I can solve and insights I can uncover with various datasets. My employement experiences have led me to work with data in supply chain, marketing, manufacturing, HR, marketing, and other areas of business. I am open to remote Data Scientist and Machine Learning Engineer positions in a variety of industries.

Outside of work I am usually busy with sports. I coach high school football in Arcadia, Wisconsin and AAU basketball in the same area. This last year we had 6 different age groups for both boys and girls that I helped coach. My interests for side projects involve sports data (football and basketball mostly), music, and video games.

Experience

Work Experience

Data Scientist Peer-Reviewed Dermatology Research — July 2022-Present

  • Lead end-to-end statistical and machine learning analyses for peer-reviewed clinical research, partnering with physicians to define hypotheses, design studies, and translate clinical questions into structured analytical frameworks
  • Developed and validated predictive and statistical models using real-world healthcare data from multiple clinical studies comprising hundreds of thousands of patient records, applying generalized linear models, survival analysis, and zero-inflated/negative binomial models to investigate rare and imbalanced outcomes
  • Built scalable, reproducible Python and R data pipelines that automated data cleaning, transformation, and feature engineering for high-dimensional clinical datasets, enabling efficient downstream statistical modeling
  • Applied rigorous model validation techniques, including cross-validation and sensitivity analyses, to evaluate predictive performance, improve model robustness, and ensure reliable statistical inference across observational healthcare datasets
  • Conducted survival and time-to-event analyses to evaluate treatment effectiveness across patient cohorts while controlling for confounding factors
  • Translated complex model outputs into clinically interpretable metrics, including odds ratios, incidence rate ratios, and survival curves, enabling publication-ready research and data-driven clinical decision-making
  • Led statistical and machine learning analyses across multiple clinical studies of rare and understudied cancer types, resulting in 2 peer-reviewed publications in JAAD (Journal of the American Academy of Dermatology)
  • Presented statistical findings through clear visualizations and structured reporting to physicians and research collaborators, translating complex analyses into actionable clinical insights
  • Maintained version-controlled analytical workflows using Git to ensure reproducibility, collaboration, and scalable research across multiple clinical studies

Data Scientist Mattress Firm — November 2021-July 2022

  • Developed and deployed an XGBoost-based customer lifetime value (CLV) model to production on Google Cloud Platform (BigQuery ML, Vertex AI), achieving 82% predictive performance and running on a monthly automated retraining and scoring cadence to support ongoing customer retention initiatives
  • Developed and deployed a K-means clustering model on GCP to segment more than 50 million customer records, enabling targeted marketing initiatives and data-driven customer segmentation
  • Applied Principal Component Analysis (PCA) to reduce 55 input variables to 7-8 key components ahead of clustering, addressing multicollinearity and improving segment interpretability
  • Published model outputs to BigQuery tables consumed by Power BI/Tableau dashboards and used in analyses by business and analytics teams, enabling self-service access to CLV scores and customer segments
  • Monitored deployed model performance over time to support timely retraining and troubleshooting decisions
  • Built scalable SQL and Python data pipelines that automated the extraction, transformation, and integration of large-scale customer and sales datasets for analytics, feature engineering, and machine learning model development
  • Designed Python (NLTK) NLP pipelines to perform sentiment analysis and extract structured insights from unstructured customer feedback, improving customer experience analysis and operational decision-making
  • Partnered with marketing, analytics, and business leadership to translate analytical insights into machine learning solutions and strategic decision-making
  • Supported experimentation and causal inference initiatives, contributing to A/B testing and media mix modeling analyses used to evaluate marketing performance
  • Analyzed macroeconomic conditions and stimulus policy impacts on sales performance, identifying revenue trends associated with external economic events to support business planning
  • Integrated Git-based version control into analytics workflows to improve reproducibility and streamline model development

Quantitative Analyst NC State Baseball — March 2021-June 2022

  • Analyzed player and pitcher performance using statistical methods, R, and Tableau to generate predictive insights supporting coaching strategy and game planning
  • Developed statistical performance metrics and interactive visualizations to support player evaluation, in-game decision-making, and player development
  • Delivered data-driven recommendations to coaching staff that informed tactical decisions and contributed to the program's College World Series semifinal appearance

Data Analyst/BI Analyst November 2019-November 2021

  • Automated SQL-based data extraction and reporting workflows, improving data availability while reducing manual reporting effort by 45 hours per week
  • Developed and maintained 17 enterprise Power BI dashboards supporting supply chain, finance, manufacturing, and HR operations, enabling KPI reporting and organization-wide data-driven decision-making
  • Reduced advertising costs by approximately $300K by developing statistical models in R to evaluate marketing spend effectiveness and identify optimization opportunities
  • Reduced logistics planning time by 8 hours per week through the development of a Python-based optimization tool leveraging real-time inventory and port data for operational decision-making
  • Partnered with cross-functional stakeholders across supply chain, finance, manufacturing, and HR to translate operational challenges into scalable analytics and reporting solutions
Education

Education

Master of Science, Statistics

North Carolina State University
  • Clubs: AI, Sports Analytics
  • Relevant Courses: Deep Learning with Neural Networks, Statisical Learning, Fundamentals of Regression, Analysis of Big Data, Data Science for Statistics, Fundamentals of Statistical Inference (I and II), Applied Statistical Methods (I and II)

Bachelor of Science, Mathematics

Xavier Univeristy
  • Concentration in Economics
  • Clubs: Clubs: Math Club, Economics Association, Habitat for Humanity, Relay for Life
  • Relevant courses: Econometrics, Probability Theory, Public Economics, Discrete Mathematics, Differential Equations

What I do?

Some areas of expertise

Machine Learning

Classification, Regression, Clustering, Deep Learning, NLP, Rec Systems

Programming

Python, R, PyTorch, pandas, NLTK, Tidyverse, numpy, sklearn, PySpark

Tools

Power BI, Tableau, R Shiny, GitHub, Jupyter Notebook, RStudio, Google Colab

Data Engineering

SQL, T-SQL, MySQL, BigQuery

Cloud Technologies

GCP, Docker

Other

Research, Presentation, Communication, Critical Thinking, Problem Solving

Publications

Publications

My Projects

Recent Work

Terrain Identification

Use deep learning techniques to identify terrain types for a prosthetic limb based on accelerometer and gyroscope readings.
  • PyTorch
  • Deep Learning
  • CNN

Emotion Detection

Use deep learning techniques to classifiy human emotion behind a conversation.
  • PyTorch
  • Deep Learning
  • CNN
  • NLP

NFL Win Prediction

An R shiny app that allows a user to explore NFL game data, and create models with selected parameters to predict winners for selected weeks.
  • R
  • R Shiny
  • Classification

Shiny App

Mobile Game User Retention A/B Testing

This project conducts an A/B Test for user retention in the mobile game Cookie Cats using the following methods: a bootstrap, Chi-Squared Test, and a Bayesian Test.
  • A/B Testing
  • Python

Music Recommendation System with ALS

This project creates a music recommendation system using Alternating Least Squares that with suggest music artists to a user based on their listening history and collaborative filtering.
  • Recommendation System
  • PySpark
  • Python

Alzheimers Risk Factors

Explore MRI scan data of demented and non-demented individuals to identify risk factors associated with Alzheimer-onset dementia
  • PySpark
  • Python
  • Classification

Motorcycle Sales

Explore motorcycle sale data to implement both the Grid Search Method and Gradient Descent Method to predict the selling price with the lowest variance.
  • Python
  • pandas
  • matplotlib

Data Channel Prediction

Analyze an online news popularity dataset to provide a report for 6 different data channels and predict the number of shares.
  • R
  • Regression

Covid API

A vignette that shows how to retrieve data from the Covid API for exploratory analysis.
  • R
  • Tidyverse
  • API

Red Wine Quality

Analyze the quality of red wine using various regression and classification models.
  • R
  • Classification
  • Regression

Big Data Project

This project provides a synopsis of Netflix's big data pipeline, investigates data in a sample database with SQL, and summarizes NFL data using the Pandas API on Spark.
  • PySpark
  • SQL
  • pandas

Get in Touch

Contact

Interested in working together or have a role that I may be a fit for? Feel free to reach out!

xgenelin@gmail.com

(507) 459-0673