Zhankui Lu
Senior Machine Learning Engineer at Amazon
Bellevue, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolZhankui Lu is a Senior Machine Learning Engineer with 10 years of experience building and scaling ML systems, currently accelerating training of large foundation models at Amazon in Bellevue. He brings a deep engineering background from AWS where he contributed to SageMaker tooling and pipelines—including adding profiler and experiment configuration support to the widely used sagemaker-python-sdk—and a stint optimizing LLM inference on AWS Neuron at Annapurna Labs. With a Ph.D. in Astrophysics from UMass Amherst, Zhankui combines rigorous research methodology with pragmatic production engineering to optimize performance across hardware and software stacks. He excels at turning complex model-training workflows into reliable, scalable pipelines and improving developer experience through documentation and examples. Known for bridging low-level inference acceleration and high-level ML platform features, he routinely navigates trade-offs between speed, cost, and reproducibility.
10 years of coding experience
16 years of employment as a software developer
Bachelor of Science (BS), Astrophysics, Bachelor of Science (BS), Astrophysics at Nanjing UniversityDoctor of Philosophy (Ph.D.), Astrophysics, Doctor of Philosophy (Ph.D.), Astrophysics at University of Massachusetts Amherst
Chinese, English
Github Skills (8)
machine-learning10
aws10
python10
pipeline10
sagemaker10
pytorch8
tensorflow8
documentation7
Programming languages (5)
Github contributions (5)
aws/sagemaker-python-sdk
Mar 2021 - Aug 2022
A library for training and deploying machine learning models on Amazon SageMaker
Role in this project:
ML Engineer
ML EngineerContributions:99 reviews, 5 commits, 10 PRs in 1 year 5 months
Contributions summary:Zhankui contributed to the `aws/sagemaker-python-sdk` repository by implementing and modifying SageMaker pipeline components. Their work included adding support for profiler configuration within the pipeline training job step. They also added the pipeline experiment configuration. Further improvements were made to documentation and code examples related to pipeline functionality.
pytorchsagemakerdeployingmxnetpython
MyOrganztion/MyRepo
Sep 2016 - Feb 2023
Contributions:1 push in 6 years 5 months
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.