I am a data scientist and a PhD student at DePaul University. I am doing research on the intersection between data science and information warfare. My specialties include computational linguistics, machine learning, creative applications for factor analyses, and web-scraping.

Outside of research, I work in the automitive/trucking industry, and have industry experience in utilities and medical devices/healthcare.

Packages

I am building an ecosystem of efficiency ML tools that allow Data Scientists to get faster wins. This means easier data collection, automating the science part of ML (logging the model results, splitting the dataset, and building baseline models), as well as proving an easy interface for using newer machine learning models that have not become mainstream yet (or have been forgotten in the back of an older textbooks).

I am actively developing three tools:

• Bukka - A ML project framework (inspired by django) that automates the boring stuff of setting up reproducibility and infrastructure to enable data scientists to quickly get started on doing the fun parts of the problem as efficiently as possible.
• NandorAPI - Efficient tools for helping to quickly build an API client on an arbitrary API.
• ExplodingHam - A package of overlooked sklearn-compatible machine learning models, focusing especially on performant small models.

Scripts

Here are a few older projects I kept here: List of published scripts

Publications

List of academic papers