Felix Abecassis
Senior Systems Software Engineer Deep Learning Performance Engineer at NVIDIA
California, United States
Summary
🤩
Rockstar🎓
Top SchoolFelix Abecassis is a senior systems software engineer and deep learning performance engineer with 13 years of experience designing GPU-aware container and orchestration tooling at NVIDIA. He maintains high-impact open-source projects like nvidia-docker and pyxis, has contributed to Kubernetes device plugin design, and helped bring GPU support to Mesos and Docker ecosystems. His work spans low-level systems, runtime and DevOps engineering—fixing memory leaks, tightening error handling, and improving container runtimes and GPU device management. He also has hands-on experience in multimedia and mobile (VLC/Android) and in accelerating DL frameworks and inference servers, demonstrating a rare mix of performance optimization and production-grade tooling. Based in California, he combines research-grade HPC training with practical contributions that enable GPU-accelerated workloads at scale.
13 years of coding experience
4 years of employment as a software developer
Master's degree, High Performance Computing, Master's degree, High Performance Computing at Ecole Centrale ParisComputer Science, Computer Science at UdacityMaster of Science, Computer Science, Master of Science, Computer Science at EPITA: Ingénierie Informatique
Github Skills (67)
python10
testing10
test-framework10
bash10
c1110
render10
c1710
javas10
system10
deep-learning10
videojs10
cudnn10
system-programming10
hardware-acceleration10
dockers10
Programming languages (16)
Github contributions (5)
NVIDIA/nvidia-docker
Nov 2015 - Nov 2018
Build and run Docker containers leveraging NVIDIA GPUs
Role in this project:
Back-end & DevOps Engineer
Back-end & DevOps EngineerContributions:2 releases, 223 commits, 37 PRs in 3 years
Contributions summary:Felix primarily contributed to the development and enhancement of the NVIDIA Docker plugin. They implemented JSON output for the `/gpu/status` and `/gpu/info` endpoints, adding functionality to retrieve GPU information in a structured format. Further contributions involved adding a Mesos agent CLI endpoint, and fixing build processes. The user also made improvements to error handling, refactoring, and Docker compatibility.
cudanvidia-dockercontainersnvidialeveraging
NVIDIA/nvidia-container-runtime
Sep 2017 - Nov 2018
NVIDIA container runtime
Role in this project:DevOps Engineer
DevOps EngineerContributions:7 releases, 119 commits, 6 PRs in 1 year 2 months
Contributions summary:Felix's contributions primarily center around enhancing the NVIDIA container runtime project's build and configuration processes. They implemented features to support various runc versions, updated the project to use the latest runtime-spec, and introduced generic runtime requirement checks by implementing environment variables. Furthermore, the user refactored the configuration, incorporated support for Docker Swarm generic resources, and modified the build process for command-line interface configurations.
nvidiagpuruntimedockercontainer-runtime
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.