Mat Leonard
Data Scientist
San Francisco, California, United States
Summary
🤩
RockstarMat Leonard is a San Francisco–based data scientist with 13 years of experience building practical ML systems and educational content around AI and data science. He blends hands-on engineering—contributions to PySyft and deep-learning PyTorch projects—with a passion for teaching, having authored notebooks and course material for Udacity and Kaggle. His open-source work ranges from image classifiers and sentiment-analysis RNNs to implementing custom autograd logic, showing comfort across applied ML and lower-level algorithmic tooling. Formerly teaching at Udacity and Kaggle, he uniquely combines curriculum design with production-minded model development, making complex concepts accessible while keeping an eye on reproducibility and gradient correctness.
13 years of coding experience
Github Skills (34)
pytorch10
operation10
python10
data-science10
random-forest-classifier10
image-classification10
classification10
scikit10
tensorrt10
machine-learning10
rnn-model10
n10
deep-learning10
tensorflow10
federated-learning10
Programming languages (8)
Github contributions (5)
udacity/deep-learning
Jan 2017 - Jan 2018
Repo for the Deep Learning Nanodegree Foundations program.
Role in this project:
ML Engineer & Data Scientist
ML Engineer & Data ScientistContributions:124 commits, 63 PRs, 109 pushes in 11 months
Contributions summary:Mat contributed to the deep learning foundations program by adding sentiment analysis notebooks and updating content. The primary focus of their work appears to be on introducing and explaining sentiment analysis models with RNNs and exploring transfer learning techniques. The user also updated images and provided additional notebook material.
deep-learning
udacity/aipnd-project
Feb 2018 - Apr 2019
Project code for AI Programming with Python ND Program
Role in this project:ML Engineer
ML EngineerContributions:10 commits, 3 pushes in 1 year 1 month
Contributions summary:Mat's primary contribution is focused on developing an image classifier project, starting with the initial notebook and progressively improving it based on user feedback. They made iterative changes to the notebook, including fixing typos, correcting label mappings, and adding notes regarding training and checkpoint saving. The changes indicate an active involvement in the project's development, refinement, and ensuring clarity for other users.
aipython
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.