Colin Raffel
Old Toronto, Ontario, United States
Summary
🤩
Rockstar🎓
Top SchoolColin Raffel is a research-oriented machine learning engineer with 14 years of experience specializing in sequence models, audio/music signal processing, and practical tooling for ML research. He holds a Ph.D. in Electrical and Electronics Engineering and an MS in Music, Science and Technology, blending rigorous theory with hands-on audio and ML engineering. Colin has contributed core functionality to widely used open-source projects—such as librosa and pretty-midi for music analysis and T5-related repositories and Mesh TensorFlow for large-scale transformer work—demonstrating both domain depth and system-level thinking. His contributions span algorithm design, performance optimizations, evaluation pipelines, and GPU-focused implementations, reflecting a knack for turning research ideas into production-quality code. Based in Old Toronto, he combines academic rigor with prolific open-source impact across music information retrieval and modern NLP toolchains.
15 years of coding experience
Doctor of Philosophy (Ph.D.), Electrical and Electronics Engineering, Doctor of Philosophy (Ph.D.), Electrical and Electronics Engineering at Columbia University in the City of New YorkBachelor of Arts, Mathematics, Physics, Bachelor of Arts, Mathematics, Physics at Oberlin CollegeMaster's degree, Music, Science and Technology, Master's degree, Music, Science and Technology at Stanford University
Github Skills (44)
algorithms10
scipy10
dspic10
python10
evaluation10
data-set10
pandas10
gpu-programming10
machine-learning10
audio-analysis10
numpy10
datasets10
midi10
music-information-retrieval10
midi-instrument10
Programming languages (7)
Github contributions (5)
mir-evaluation/mir_eval
Aug 2013 - Apr 2022
Evaluation functions for music/audio information retrieval/signal processing algorithms.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:6 releases, 17 reviews, 479 commits in 8 years 9 months
Contributions summary:Colin's contributions primarily involved implementing and refining evaluation functions for music and audio information retrieval algorithms. Their commits focused on developing metrics for beat tracking and onset detection, demonstrating a strong understanding of the project's core domain. The user translated and refactored code, fixed bugs, and ensured their functions worked as intended within the project's evaluation framework.
audioevaluation-functioninformation-retrievalsignal-processing
craffel/pretty-midi
Aug 2013 - Feb 2022
Utility functions for handling MIDI data in a nice/intuitive way.
Role in this project:Back-end Developer
Back-end DeveloperContributions:13 releases, 21 reviews, 231 commits in 8 years 8 months
Contributions summary:Colin primarily worked on implementing the core functionality for the pretty-midi library. They focused on adding and refining features like piano roll generation, chroma calculations, and synthesis methods. Their work involved the addition of instrument and note containers, as well as code to handle and incorporate pitch bends and control changes. The user also implemented functionality for MIDI file writing and beat tracking.
midi
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.