Chris Donahue
Assistant Professor at Google DeepMind
Pittsburgh, Pennsylvania, United States
Summary
🤩
Rockstar🎓
Top SchoolChris Donahue is an assistant professor at Carnegie Mellon and part-time research scientist at Google Magenta with 13 years of experience at the intersection of machine learning, music, and creative interaction. He builds and evaluates generative audio models—contributing to notable projects like WaveGAN and Magenta.js where he implemented Piano Genie enhancements and audio evaluation metrics such as inception scores. His background spans academic research (PhD in Music, postdoc at Stanford) and industry research roles at Google DeepMind, giving him a rare blend of musical expertise and ML rigor. Chris is especially skilled at making ML models more usable for creative practitioners, from browser-based music generation to raw audio synthesis.
13 years of coding experience
1 year of employment as a software developer
University of California, San DiegoBachelor of Science - BS Computer Science, Bachelor of Science - BS Computer Science at The University of Texas at AustinPostdoctoral Scholar Computer Science, Postdoctoral Scholar Computer Science at Stanford University
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (17)
machine-learning-models10
evaluation10
model-driven10
tensorflowjs10
machine-learning10
rnn-model10
audio-processing10
n10
generative-adversarial-network10
model-building10
eval10
lstm10
tensorflow10
modeling10
model-driven-development10
Programming languages (11)
Github contributions (5)
chrisdonahue/wavegan
Feb 2018 - Nov 2022
WaveGAN: Learn to synthesize raw audio with generative adversarial networks
Role in this project:
ML Engineer
ML EngineerContributions:101 commits, 6 PRs, 27 pushes in 4 years 10 months
Contributions summary:Chris's contributions focused on adding and implementing evaluation metrics for a WaveGAN model, a generative adversarial network for audio synthesis. This included incorporating the inception score to assess the quality of generated audio samples. The user worked with TensorFlow and potentially related libraries like librosa, demonstrating expertise in machine learning model evaluation and audio processing.
pytorchdeep-learningadversarialaudiogenerative-adversarial-networks
magenta/magenta-js
Nov 2018 - Aug 2019
Magenta.js: Music and Art Generation with Machine Learning in the browser
Role in this project:ML Engineer
ML EngineerContributions:5 commits, 7 PRs, 23 comments in 8 months
Contributions summary:Chris's contributions primarily revolve around enhancing the Piano Genie model within the magenta-js repository. They implemented the Piano Genie model, refactored the API for improved clarity, and extended its functionality with new features like conditioning on key signatures and chord families. Further improvements included adding a model warmup during initialization. The user appears to be a core contributor to this specific ML-driven project.
artbrowsermagenta-jsjavascriptmagenta
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.