Clément Chadebec
Research Scientist at Jasper
Paris, Ile-de-France
Summary
🤩
Rockstar🎓
Top SchoolClément Chadebec is a research scientist based in Paris with seven years of experience at the intersection of deep generative models and applied mathematics. Currently at Jasper (after a stint at Stability AI), he focuses on building and accelerating diffusion, flow, and bridge models and leads open-source releases that bridge research and engineering. His PhD and prior work at Inria explored geometric and Riemannian aspects of generative models, giving him a strong theoretical footing that informs practical model improvements. An active practitioner in PyTorch, he contributed unified VAE implementations (including RHVAE, IWAE, Beta-TCVAE and linear normalizing flows) that support reproducible experiments and benchmarks. Colleagues value him for combining rigorous math training with hands-on engineering to make state-of-the-art generative models faster and more accessible.
7 years of coding experience
1 year of employment as a software developer
Master of Science - MS Machine Learning & Applied Mathematics (MVA), Master of Science - MS Machine Learning & Applied Mathematics (MVA) at ENS Paris-SaclayBachelor of Science (B.Sc.) Mathematics & Physics, Bachelor of Science (B.Sc.) Mathematics & Physics at Lycée Condorcet, ParisDoctor of Philosophy - PhD Mathematics and Computer Science, Doctor of Philosophy - PhD Mathematics and Computer Science at Université Paris CitéMaster's degree Science and Executive Engineering Applied Mathematics, Master's degree Science and Executive Engineering Applied Mathematics at Mines Paris - PSL
French, English, Spanish
Github Skills (4)
machine-learning10
autoencoder10
deep-learning10
pytorch10
Programming languages (2)
Github contributions (5)
clementchadebec/benchmark_VAE
Oct 2021 - Dec 2022
Unifying Variational Autoencoder (VAE) implementations in Pytorch (NeurIPS 2022)
Role in this project:
ML Engineer
ML EngineerContributions:12 releases, 69 reviews, 444 commits in 1 year 2 months
Contributions summary:Clément primarily contributed to the development and testing of Variational Autoencoder (VAE) models within the PyTorch framework. They implemented new models such as the Residual Hyperbolic VAE (RHVAE), the IWAE, the Beta-TCVAE, and a VAE with a linear normalizing flow. Furthermore, they have also created and tested example notebooks and training scripts showcasing their implementations.
benchmarkingvaevariational-inferencevae-implementationnormalizing-flows
louis-j-vincent/benchmark_VAE
Feb 2022 - Oct 2022
Library for Variational Autoencoder benchmarking
Contributions:65 commits, 2 PRs, 4 pushes in 8 months
autoencoderbenchmarkingbenchmarkvariational-autoencodervariational
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.