Evan Lezar
Senior System Software Engineer at NVIDIA
Berlin, Germany
Summary
🤩
Rockstar🎓
Top SchoolEvan Lezar is a Senior System Software Engineer based in Berlin with 15 years of professional experience and a PhD-grounded background in computational electromagnetics. He specializes in low-level systems, container runtimes, and DevOps automation, contributing significantly to high-profile open-source projects at NVIDIA, Moby, and Podman where he improved build systems, CDI/device integration, and CI/CD testing for GPU workloads. At NVIDIA he drives production-ready tooling for GPU containers and Kubernetes device plugins, blending deep systems programming with practical deployment engineering. His career spans major cloud and infrastructure teams at AWS and Amazon, and he’s known for refactoring complex codebases and making build and release pipelines portable across architectures. Colleagues describe him as adaptable and persistent—equally comfortable translating academic research into robust production systems and untangling cross-platform build issues that block releases.
15 years of coding experience
18 years of employment as a software developer
BSc, Physics, Mathematics, Computer Science, Distinction, BSc, Physics, Mathematics, Computer Science, Distinction at Stellenbosch University
English, Afrikaans, German, Japanese
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (45)
dependency-management10
go-modules10
kubernetes10
gnu-make10
container10
docker10
scripting10
command-line-interface10
linux10
makefile10
containerization10
dockers10
cicd10
automation10
occi10
Programming languages (19)
Github contributions (5)
NVIDIA/nvidia-container-toolkit
Jun 2021 - Jan 2023
Build and run containers leveraging NVIDIA GPUs
Role in this project:
Back-end & DevOps Engineer
Back-end & DevOps EngineerContributions:47 releases, 561 reviews, 153 commits in 1 year 7 months
Contributions summary:Evan's contributions centered on enhancing the `nvidia/nvidia-container-toolkit` project. Their work included refactoring command-line arguments by moving runtime flags to a struct. Further contributions included adding the ability to specify the NVIDIA Container Toolkit CLI path. The changes suggest involvement in both back-end code changes to make the toolkit function correctly. They also included adding CI/CD features to make this buildable and releaseable, showing DevOps skills.
containersnvidialeveraginggpudocker
NVIDIA/k8s-device-plugin
Jan 2021 - Nov 2022
NVIDIA device plugin for Kubernetes
Role in this project:Back-end & DevOps Engineer
Back-end & DevOps EngineerContributions:19 releases, 740 reviews, 151 commits in 1 year 10 months
Contributions summary:Evan primarily contributed to the NVIDIA device plugin's back-end infrastructure and configurations. They implemented features related to handling and exposing MIG devices, including their profiles and health checks, utilizing the go-nvml library. Furthermore, the user refactored plugin construction, integrated CLI tools for improved deployment, and added support for GDS and MOFED. These changes focused on improving the plugin's usability, robustness, and compatibility with various NVIDIA GPU setups.
nvidia-dockernvidiagpukubernetesdevice-plugin
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.