Jason Yosinski
Member Of Technical Staff at ML Collective
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolJason Yosinski is an ML research scientist and entrepreneur with 15+ years of experience probing how neural networks learn and fail, currently a Member of Technical Staff at OpenAI and co-founder of the nonprofit ML Collective. He co-founded and led research teams at Uber AI Labs and startups focused on applying ML to real-world systems, including Windscape AI for wind energy optimization. Jason combines deep academic roots (PhD work at Cornell Creative Machines Lab) with hands-on engineering—authoring and maintaining influential open-source tools like the deep-visualization-toolbox for understanding neural activations. His background spans full-stack implementation, research, and teaching, from C++/Matlab research code to UI-driven visualization improvements in widely used repos. Known for making research accessible, he actively supports open collaboration and education in ML while advising early-stage technical ventures. An offbeat but revealing detail: he historically prefers Twitter/email over LinkedIn for technical conversation, reflecting a bias toward direct, real-time engagement.
15 years of coding experience
8 years of employment as a software developer
study abroad Engineering, study abroad Engineering at University of CambridgeCalifornia Institute of TechnologyPh.D. Computer Science, Ph.D. Computer Science at Cornell University
spanish (poorly)
Github Skills (11)
computer-vision10
machine-learning10
caffe10
deep-learning10
python10
numpy9
documentation9
opencv9
unit-testing9
user-interface9
neural-network8
Programming languages (8)
Github contributions (5)
yosinski/deep-visualization-toolbox
Apr 2015 - Sep 2017
DeepVis Toolbox
Role in this project:
Full-stack Developer
Full-stack DeveloperContributions:120 commits, 8 PRs, 67 pushes in 2 years 5 months
Contributions summary:Jason primarily worked on the `caffevis/app.py` and related files, implementing core functionalities for a deep learning visualization toolbox. They focused on features within the CaffeVisApp, including image loading, unit visualization, and key binding improvements. The commits also involved updating the settings file and modifying the user interface to improve usability and provide a better user experience.
deep-learningtoolbox
lisa-lab/pylearn2
Sep 2011 - Feb 2014
Warning: This project does not have any current developer. See bellow.
Role in this project:ML Engineer
ML EngineerContributions:13 commits in 2 years 5 months
Contributions summary:Jason primarily contributed to code maintenance, bug fixes, and documentation updates within the pylearn2 repository. They addressed a type checking bug related to CudaNdarray and updated code to improve consistency. The user also corrected documentation errors and updated tutorials and added documentation to unit tests. Finally, they fixed a bug related to base class initialization within the MLP module.
javascripttypescript
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.