Skills
The tools and methods I use to take retail data problems from question to production.
Programming Languages
Python for data science and production machine learning, SQL for everything data, plus R for statistics and JavaScript / Node.js for building data apps.
Data Analysis & Statistics
Cleaning, reshaping, and feature engineering with Pandas and NumPy, plus the exploratory analysis, statistical modeling, and hypothesis testing that keep conclusions honest.
Machine Learning & AI
Classification, regression, recommendation, and forecasting with scikit-learn, XGBoost, and TensorFlow, plus LLM-powered and agentic AI workflows.
Forecasting & Optimization
Time-series and machine learning forecasters, from ARIMA to gradient boosting, turned into demand plans, inventory targets, and allocation decisions.
Data Visualization & Apps
Seaborn, Matplotlib, and Plotly for analysis, Folium and GeoPandas for maps, and award-winning Streamlit apps so stakeholders can explore the work themselves.
Cloud & MLOps
Building and deploying pipelines on GCP and AWS, with Databricks and Spark for scale and Kubeflow plus CI/CD to keep them reproducible and monitored.
Databases
SQL across MySQL, PostgreSQL, and BigQuery for analytical and production workloads, and MongoDB for document data.
Version Control & Collaboration
Git and GitHub day to day for branching, code review, and shipping readable, well-documented code a team can maintain. I also teach these practices at the University of Denver.