Kevin Klues
Distinguished Engineer at NVIDIA
Berlin, Germany
Summary
🤩
Rockstar🎓
Top SchoolKevin Klues is a Distinguished Engineer with 18 years building production-grade systems and cloud-native infrastructure, currently leading systems software work at NVIDIA from Berlin. His career blends deep academic roots (PhD work at UC Berkeley) with hands-on impact across Kubernetes, Mesos, and DC/OS—contributions that span CPU/NUMA resource management, GPU device management and NVML/DCGM bindings, and robust test automation. Known for shipping stable build and cross-architecture runtime tooling, he has driven improvements that prevent crashes on older drivers and enabled MIG and time-slicing support in containerized GPU environments. Kevin routinely bridges low-level device interfaces and large distributed platforms, pairing algorithmic improvements with pragmatic DevOps and CI/test fixes. An uncommon mix of research rigor and production pragmatism, he’s also active in major open-source projects that underpin modern datacenter orchestration.
18 years of coding experience
20 years of employment as a software developer
Ph.D Computer Science, Ph.D Computer Science at University of California, BerkeleyTelecommunications, Telecommunications at Technische Universität BerlinM.S. Computer Engineering, M.S. Computer Engineering at Washington University in St. LouisB.S. Electrical Engineering / Computer Science, B.S. Electrical Engineering / Computer Science at Rose-Hulman Institute of Technology
English, German, Italian, Spanish
Stackoverflow
877reputation
1.2mreached
19answers
7questions
Github Skills (53)
mesos10
kubernetes10
docker10
gnu-make10
monitor10
python10
crossbuild10
testing10
linux10
bash10
c1110
makefile10
dockers10
dcos10
c1710
Programming languages (14)
Github contributions (5)
NVIDIA/k8s-device-plugin
Jan 2019 - Nov 2022
NVIDIA device plugin for Kubernetes
Role in this project:
Back-end Developer
Back-end DeveloperContributions:29 releases, 263 reviews, 500 commits in 3 years 10 months
Contributions summary:Kevin primarily contributed to the NVIDIA device plugin for Kubernetes by implementing and improving core functionalities related to GPU device management and health checking. Their contributions included refactoring and refactoring the health check path for the "watchXIDs" process and also added the functionality to optionally pass back a list of device node specifications for the kubelet for handling time-slicing and replica requests. They also addressed potential bugs by changing or making the application's behavior in ways that prevent crashes when used in older driver releases, and also added various code formatting and refactoring fixes.
nvidia-dockernvidiagpukubernetesdevice-plugin
NVIDIA/libnvidia-container
Feb 2019 - Mar 2022
NVIDIA container runtime library
Role in this project:Backend & DevOps Engineer
Backend & DevOps EngineerContributions:16 releases, 10 reviews, 274 commits in 3 years 2 months
Contributions summary:Kevin primarily contributed to the build system and infrastructure of the NVIDIA container runtime library. They reworked the build system to support cross-building for multiple architectures and added support for various operating systems and architectures. Additionally, the user implemented enhancements related to device access within containers, including support for MIG devices and enabling the injection of capabilities files.
nvidiagpuruntimedockerruntime-library
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.