Vincent Lostanlen
Research Scientist at Centre national de la recherche scientifique
Nantes, Pays de la Loire, France
Summary
🤩
Rockstar🎓
Top SchoolVincent Lostanlen is a CNRS research scientist and former NYU visiting scholar who builds machine listening systems that span music, bioacoustics, and urban soundscapes. With 11 years of experience and a PhD from École normale supérieure, he leads multiple interdisciplinary projects (from multi-resolution neural nets to bioacoustic conservation) and recruits PhD students and postdocs in AI for music and wildlife monitoring. He combines theory—closed-form sound models and scattering transforms—with applied work, contributing to GPU-accelerated open-source tools like Kymatio to improve performance and testing for real-world ML. Based in Nantes, he bridges sensor design, self-supervised learning, and ecological monitoring, and brings an uncommon blend of musical training (specialized jazz study) and rigorous signal-processing expertise.
11 years of coding experience
4 years of employment as a software developer
Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at Ecole normale supérieureMaster of Science - MS, Computer Science, Master of Science - MS, Computer Science at Sorbonne UniversitéCycle spécialisé, Jazz, Cycle spécialisé, Jazz at Conservatoire à rayonnement régional de Paris - Ida RubinsteinEngineer's degree, Computer Science, Engineer's degree, Computer Science at Télécom Paris
Stackoverflow
101reputation
0reached
0answers
0questions
Github Skills (12)
keras10
pytorch10
benchmarking10
benchmark10
deep-learning10
gpu-acceleration10
python10
testing10
numpy9
tensorflow9
wavelet9
machine-learning9
Programming languages (7)
Github contributions (5)
kymatio/kymatio
Feb 2021 - Jan 2023
Wavelet scattering transforms in Python with GPU acceleration
Role in this project:
Back-end Developer & MLOps Engineer
Back-end Developer & MLOps EngineerContributions:196 reviews, 89 commits, 154 PRs in 1 year 11 months
Contributions summary:Vincent contributed to implementing and improving the Kymatio library, focusing on the Keras and PyTorch backends. They implemented the `get_config` method for Keras layers. Additionally, the user added and improved benchmarks for various backends including torch, and improved tests of 1D, 2D, and 3D scattering. These changes indicate a focus on testing and performance enhancements, particularly for GPU acceleration.
pytorchwaveletswaveletpythongpu-acceleration
lostanlen/scattering.m
Mar 2021 - May 2022
Contributions:2 releases, 33 commits, 74 PRs in 1 year 2 months
matlab-toolboxsignalmatlabtoolboxscattering
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.