Jean-baptiste Keck
Porteur De Projet RayMapr at Université Grenoble Alpes
Grenoble, Auvergne-Rhône-Alpes, France
Summary
👤
Senior🎓
Top SchoolJean-baptiste Keck is a computational scientist and project lead at Université Grenoble Alpes with 12 years of experience in high-performance numerical simulation, multi-GPU distributed computing, and computational imaging. He holds a PhD in applied mathematics and computer science and has developed production-grade libraries for hybrid distributed computing, on-the-fly OpenCL code generation and automated architecture-specific optimization. His recent work combines inverse problem techniques, optimal transport and autodiff for optical device design, alongside large-scale multi-GPU implementations. An active open-source contributor, he has improved CuPy’s kernel argument handling to support structured types and added tests and documentation for GPU kernels. Based in Grenoble, he pairs deep research abilities with practical cluster administration and teaching experience, making him adept at translating advanced numerical methods into efficient, deployable software.
12 years of coding experience
Doctor of Philosophy - PhD, Mathématiques et informatique appliquées, Doctor of Philosophy - PhD, Mathématiques et informatique appliquées at Université Grenoble AlpesEcole Nationale Supérieure d'Informatique et de Mathématiques Appliquées de GrenobleClasse Préparatoire aux Grandes Ecoles (équivalence L2), Mathématiques, Physiques, Sciences de l'Ingénieur, Classe Préparatoire aux Grandes Ecoles (équivalence L2), Mathématiques, Physiques, Sciences de l'Ingénieur at Lycée Kléber, Strasbourg
French, English, German
Github Skills (9)
kernel10
cuda10
python10
kernel-mode10
numpy10
testing10
cython9
cprogramming-language9
c-language9
Programming languages (6)
Github contributions (5)
cupy/cupy
Mar 2021 - Apr 2021
NumPy & SciPy for GPU
Role in this project:
Backend Developer
Backend DeveloperContributions:5 reviews, 20 commits, 2 PRs in 1 month
Contributions summary:Jean-baptiste primarily contributed to the CuPy library by implementing features related to kernel arguments. They added support for passing structured data types by value in kernel parameters, addressing a specific issue. Furthermore, the user fixed flake8 pyx errors, and added kernel argument tests to ensure functionality and code quality. Documentation was also added to cover kernel arguments.
cudapythoncusolvergpunumpy
Poulpy21/papers
Jan 2015 - Feb 2015
Contributions:46 commits in 1 month
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.