Matt Epland
AI & GenAI Engineer at Codoxo
New York, New York, United States
Summary
🤩
Rockstar🎓
Top SchoolMatt Epland is an AI & GenAI Engineer with 11 years of experience who builds production-grade generative AI products at Codoxo and brings a strong research foundation from a PhD at Duke. He combines hands-on ML engineering with practical interpretability work, contributing to popular open-source projects like imodels (FIGS) and dtreeviz to make tree-based models more transparent and better visualized. Based in New York, he bridges prototype research and production, adding features that improve developer ergonomics such as sklearn compatibility and dtreeviz integrations. Colleagues know him for pragmatic solutions that prioritize explainability and reproducibility, including tooling to extract scikit-learn trees from FIGS and clearer visual legends for decision trees.
11 years of coding experience
Doctor of Philosophy (PhD), Doctor of Philosophy (PhD) at Duke UniversityUniversity of Minnesota Twin Cities
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (15)
scikit10
data-visualizations10
decision-tree10
machine-learning10
data-visualization10
data-visualisation10
visualization10
explainable-artificial-intelligence10
visualizations10
python10
data-science10
matplotlib10
scikit-learn10
random-forest9
xgboost8
Programming languages (4)
Github contributions (5)
parrt/dtreeviz
Dec 2022 - Jan 2023
A python library for decision tree visualization and model interpretation.
Role in this project:
Data Scientist
Data ScientistContributions:43 reviews, 41 commits, 21 PRs in 25 days
Contributions summary:Matt made several contributions focused on enhancing the visualization aspects of decision trees within the dtreeviz library. Their work primarily involved fixing and updating Matplotlib warnings, improving the rendering of tree diagrams and plots, and adding new features for visualizing leaf distributions, including the addition of horizontal bar charts. They also refactored legend drawing and edge label formatting to enhance clarity and consistency.
xgboostdecision-treemodel-interpretationpythonrandom-forest
csinva/imodels
Aug 2022 - Jan 2023
Interpretable ML package 🔍 for concise, transparent, and accurate predictive modeling (sklearn-compatible).
Role in this project:Data Scientist & ML Engineer
Data Scientist & ML EngineerContributions:5 reviews, 36 commits, 9 PRs in 4 months
Contributions summary:Matt primarily contributed to the development and testing of the FIGS (Fast Interpretable Greedy-Tree Sums) model, an interpretable machine learning package. Their work included implementing features such as an optional `feature_names` parameter for printing tree structures and integrating `dtreeviz` for visualization of the model. They also created a notebook for testing FIGS and included code to extract a scikit-learn DecisionTree object from the FIGS model, indicating an effort to expand the model's functionality and integrate it with existing tools.
pythonxaisklearnsklearn-compatibleinterpretable
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.