Dong Meng
Senior Manager, Principal Engineer at NVIDIA
San Francisco Bay Area United States
Summary
👤
Senior🎓
Top SchoolDong Meng is a Senior Manager and Principal Engineer with 11 years of experience building GPU-accelerated ML and data analytics solutions on heterogeneous distributed systems. At NVIDIA he partners with hyperscalers to bring up cutting-edge GPUs (Turing, Ampere, Hopper) in cloud environments, driving cross-platform AI strategy and technical engagements that have directly generated hundreds of millions in revenue. He combines deep systems and MLOps chops—deploying Triton Inference Server to GKE, integrating RAPIDS/XGBoost into Dataproc, and tuning accelerator performance for training and inference—to deliver production-grade, high-performance ML pipelines. Trained in signal processing and compressive sensing (MS, Ohio State), he blends rigorous research foundations with practical cloud-native engineering to make accelerator performance accessible to large customers.
11 years of coding experience
14 years of employment as a software developer
Bachelor of Engineering Telecommunication Engineering Honor Class, Bachelor of Engineering Telecommunication Engineering Honor Class at Xidian UniversityMaster of Science - MS Electrical Engineering; Signal Processing/Compressive Sensing, Master of Science - MS Electrical Engineering; Signal Processing/Compressive Sensing at The Ohio State University
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (28)
kubernetes10
xgboost10
docker10
spark10
google-cloud-dataproc10
cloud-computing10
machine-learning10
bash10
inference10
dockers10
gcp10
rapids10
kubernetes-pods10
devops10
performance-monitor9
Programming languages (11)
Github contributions (5)
GoogleCloudDataproc/initialization-actions
Mar 2020 - Feb 2022
Run in all nodes of your cluster before the cluster starts - lets you customize your cluster
Role in this project:
ML Engineer
ML EngineerContributions:8 reviews, 16 commits, 32 PRs in 1 year 11 months
Contributions summary:Dong contributed to the initialization actions for Google Cloud Dataproc, specifically focusing on integrating and testing RAPIDS and XGBoost for GPU-accelerated workloads. They added support for Spark 3 GPU scheduling and updated versions of XGBoost and RAPIDS. The user also modified the installation scripts for GPU drivers and configured YARN to utilize GPU resources, demonstrating expertise in setting up and optimizing GPU-enabled environments within the Dataproc framework.
nodesgoogle-cloud-dataprockubernetescustomizeterraform
triton-inference-server/server
Mar 2021 - Jan 2023
The Triton Inference Server provides an optimized cloud and edge inferencing solution.
Role in this project:DevOps Engineer
DevOps EngineerContributions:12 reviews, 12 commits, 15 PRs in 1 year 10 months
Contributions summary:Dong primarily focused on deploying and configuring the Triton Inference Server within a GKE (Google Kubernetes Engine) environment, specifically for the GKE Marketplace. Their contributions include building deployment scripts, updating container versions, integrating benchmarking tools, and configuring performance analysis. They modified deployment scripts and updated container images, reflecting expertise in containerization, cloud deployment, and performance tuning. Additionally, the user addressed feedback to align with GKE marketplace expectations.
nvidia-dockernvidiadeep-learninggpuinference
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.