Matt Mcvicar
Machine Learning Research Manager at Apple
United Kingdom
Summary
👤
Senior🎓
Top SchoolMatt Mcvicar is a Machine Learning Research Manager at Apple Music in London with 13 years of experience building audio and music intelligence systems from research prototypes to product features. He combines deep academic training (PhD in Engineering Mathematics) and a Fulbright stint at Columbia with industry roles at Jukedeck and Apple, leading teams that power features like Apple Music Home, Sing, and AutoMix. His work spans scalable probabilistic models for harmony and large-scale audio–lyrics–social analyses, plus practical engineering contributions to open-source tooling such as librosa (notably a loudness-based chromagram for chord recognition). Comfortable bridging research and product, he focuses on foundational audio representations that reveal subtle musical styles and drive real user experiences. An often-overlooked strength is his track record in curating and mining linked music datasets to surface emerging genres and listener patterns.
13 years of coding experience
12 years of employment as a software developer
Doctor of Philosophy (Ph.D.) Engineering Mathematics, Doctor of Philosophy (Ph.D.) Engineering Mathematics at University of Bristol
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (14)
machine-learning10
audio-analysis10
python10
feature-extraction10
numpy10
signal-processing9
scikit-learn9
digital-signal-processing9
data-structure9
scikit9
algorithm9
data-structures9
algorithms9
pandas8
Programming languages (2)
Github contributions (5)
librosa/librosa
Nov 2012 - Jan 2014
Python library for audio and music analysis
Role in this project:
Back-end Developer & Data Scientist
Back-end Developer & Data ScientistContributions:38 commits, 1 comment in 1 year 2 months
Contributions summary:Matt significantly contributed to the `librosa` library by implementing and refining a loudness-based chroma feature extraction method for audio analysis. This involved the development of several functions for constant-Q transforms, Hamming window calculations, and the creation of a loudness-based chromagram. The user also integrated the new feature into the core `chromagram` function and added example scripts. Their work is focused on improving the library's audio analysis capabilities, specifically in the domain of chord recognition, with a shift from an Ellis-based method to the user's method.
python-librarydtwpythonlibrosaaudio
mattmcvicar/death2chroma
Apr 2013 - May 2013
Contributions:51 commits in 16 days
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.