Sébastien Popoff
Research Fellow (Chargé De Recherche) at CNRS - Centre national de la recherche scientifique
Paris, Ile-de-France
Summary
👤
Senior🎓
Top SchoolSébastien Popoff is a CNRS research fellow and physicist based in Paris with nine years of postdoctoral and research experience focused on optics, coherent control, wavefront shaping, and wave propagation in complex media. He leads experimental and theoretical work at the Langevin Institute, applying spatio-temporal manipulation techniques to imaging, information transmission, and characterization in scattering systems. Comfortable at the intersection of physics and computation, he has extended his toolkit into machine learning by engineering complex-valued neural network components for PyTorch to better model wave phenomena. His background spans top French institutions and international postdocs (Yale, ENS), reflecting deep expertise in time reversal and scattering that informs both fundamental studies and applied optics. Notably, he combines hands-on lab skills with open-source ML engineering, making advanced complex-valued layers accessible to the community.
9 years of coding experience
3 years of employment as a software developer
Master 2, Acoustique Physique, Master 2, Acoustique Physique at Université Denis Diderot (Paris VII) / University Paris VIIDiplôme d'Ingénieur, Physique, Chimie, Biologie, Mathématiques Appliquées, Diplôme d'Ingénieur, Physique, Chimie, Biologie, Mathématiques Appliquées at Ecole supérieure de Physique et de Chimie Industrielles de la Ville de Paris
French, English
Github Skills (9)
neural-network10
pytorch10
convolutional-neural-networks10
python9
machine-learning9
deep-learning9
relu8
dropout8
batch-normalization8
Programming languages (4)
Github contributions (5)
wavefrontshaping/complexPyTorch
May 2019 - Sep 2022
A high-level toolbox for using complex valued neural networks in PyTorch
Role in this project:
ML Engineer
ML EngineerContributions:8 releases, 28 commits, 8 PRs in 3 years 4 months
Contributions summary:Sébastien primarily contributed to the development of complex-valued neural network components within the PyTorch framework. Their work involved implementing custom layers, including complex-valued convolutional and linear layers, batch normalization, dropout, and ReLU activation functions. They also added a notebook to demonstrate the use of these complex layers, indicating an effort to make the library more accessible. The user's modifications reflect a focus on improving the library's functionality and adhering to the PyTorch coding style.
pytorchdeep-learningtoolboxhigh-levelneural-networks
wavefrontshaping/Layout
Oct 2019 - Mar 2022
A library to create layouts of various shapes for wavefrontshaping using DMDs or SLMs.
Contributions:7 releases, 42 commits, 13 PRs in 2 years 5 months
layouts
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.