David Zier
Director Of DL System Software at NVIDIA
Beaverton, Oregon, United States
Summary
🤩
Rockstar🎓
Top SchoolDavid Zier is a Director of Deep Learning System Software at NVIDIA with a decade-plus career advancing high-performance inference and profiling tooling for ML workflows. He leads the teams behind Triton Inference Server and GenAI-Perf, bringing hands-on backend engineering experience—evidenced by contributions to Triton’s custom backend integration and robust error handling. With a PhD in Electrical & Computer Engineering and a track record from CPU performance modeling to system software leadership, he blends deep architecture expertise with practical product delivery. Known for instituting rigorous engineering practices and shipping production-grade tools for cloud and edge deployment, he helps bridge research-grade performance modeling and scalable inference systems. Based in Beaverton, Oregon, he combines academic rigor (3.98 GPA PhD) with operational leadership at one of the industry’s leading AI infrastructure providers.
7 years of coding experience
18 years of employment as a software developer
Doctor of Philosophy (Ph.D.), Electrical & Computer Engineering, 3.98 / 4.00 GPA, Doctor of Philosophy (Ph.D.), Electrical & Computer Engineering, 3.98 / 4.00 GPA at Oregon State University
Github Skills (7)
error-handling10
c-language10
sdk10
cprogramming-language10
machine-learning9
inference9
gpu8
Programming languages (5)
Github contributions (5)
triton-inference-server/server
Jul 2019 - Nov 2022
The Triton Inference Server provides an optimized cloud and edge inferencing solution.
Role in this project:
Backend Developer
Backend DeveloperContributions:29 releases, 92 reviews, 202 commits in 3 years 5 months
Contributions summary:David primarily contributed to the implementation of a custom backend instance within the Triton Inference Server. Their work involved creating the `CustomInstance` class and integrating it with the identity backend, as evidenced by code additions and modifications to the custom SDK. The user also addressed code review comments, refactored and cleaned up code, and implemented error handling mechanisms, indicating a focus on backend functionality and maintainability. They interacted with custom backends, SDK, and error handling.
nvidia-dockernvidiadeep-learninggpuinference
triton-inference-server/.github
May 2022 - Oct 2022
Community health files for NVIDIA Triton
Contributions:1 review, 3 PRs, 3 pushes in 4 months
nvidiahealth-fileshealthcommunity-health-filestriton
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.