Joël Chavas
Research Scientist And Deep Learning Engineer at CEA Joliot
Orsay, Île-de-France, France
Summary
👤
Senior🎓
Top SchoolJoël Chavas is a research scientist and deep learning engineer with 12 years of experience spanning neuroscience, medical device development, embedded systems and applied ML for brain imaging. He leads research at CEA Joliot, supervising students and organizing reproducible code while applying self-supervised and supervised deep learning (PyTorch, TensorFlow) to detect neurodevelopmental biomarkers. His background includes leading system design for an IEC‑compliant, vision-restoring medical device and practical expertise in requirement engineering, traceability, FMEA and ISO 14971 risk management. Trained as a neuroscientist with a strong foundation in microelectronics and embedded firmware, he combines rigorous scientific publication practice with hands-on software engineering for regulated clinical systems. Notably, he bridges fundamental statistical‑physics research and modern deep learning workflows, enabling translational projects that move from algorithms to certified medical hardware.
12 years of coding experience
14 years of employment as a software developer
Maitrise, Chemistry, Biology, Maitrise, Chemistry, Biology at Ecole Normale Supérieure (Ulm)Master, Microelectronics, Master, Microelectronics at University of Padova (Italy)Diplomarbeit / PhD, Neuroscience, Diplomarbeit / PhD, Neuroscience at University Paris VI (France) / Göttingen (Germany)
English, German, Italian
Stackoverflow
151reputation
12kreached
19answers
3questions
Github Skills (34)
optimization6
rename6
indexing6
coordinates6
python6
dataframe6
visual-studio-code6
pandas6
matplotlib6
jupyter6
self-organizing-map5
region5
autoencoder5
hydra5
pytorch-lightning4
Programming languages (3)
Github contributions (5)
neurospin/deep_folding
Mar 2021 - Jan 2023
Contributions:28 reviews, 401 commits, 59 PRs in 1 year 10 months
neurospin-projects/2022_jchavas_cingulate_inhibitory_control
Mar 2022 - Jun 2023
Use contrastive learning to predict cognitive parameters from skeletons
Contributions:1 review, 63 pushes, 25 branches in 1 year 3 months
parameterscontrastive-learningcognitivecontrastivepredict
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.