Hongye Sun
Salt Lake City, Utah, United States
Summary
🤩
Rockstar🎓
Top SchoolHongye Sun is a final-year PhD candidate in Marketing at the University of Utah who uses field experiments, causal inference, and machine learning to solve practical salesforce and healthcare marketing problems. She collaborates with industry partners across healthcare and high-tech sectors, translating rigorous empirical work—such as studies on social ties to reduce salesforce attrition and AI/LLM applications in sales training—into actionable strategies. With seven years of experience that include lecturing, lab management, and multiple TA roles, she blends strong teaching and research communication with hands-on experimental design. Her technical toolkit spans text, image, voice, and video analysis, and she has contributed engineering improvements to prominent ML infrastructure (e.g., GPU/TPU support in Kubeflow Pipelines), reflecting a rare mix of marketing rigor and implementation fluency.
7 years of coding experience
3 years of employment as a software developer
Master of Philosophy - MPhil, Marketing, 3.8/4.0, Master of Philosophy - MPhil, Marketing, 3.8/4.0 at Nanjing UniversityThe University of UtahBachelor of Business Administration - BBA, Marketing, 3.9/4.0, Bachelor of Business Administration - BBA, Marketing, 3.9/4.0 at Jilin University
Github Skills (8)
kubeflow10
machine-learning10
python10
gcp10
kubernetes-pods9
tensorflow9
kubernetes9
mlops8
Programming languages (7)
Github contributions (5)
kubeflow/pipelines
Nov 2018 - Sep 2020
Machine Learning Pipelines for Kubeflow
Role in this project:
ML Engineer
ML EngineerContributions:10 releases, 34 reviews, 98 commits in 1 year 10 months
Contributions summary:Hongye implemented support for NVIDIA GPU resource limits, requests, and node selectors for the `ContainerOp` in the Kubeflow Pipelines SDK, enabling GPU-based training and inference. They introduced new functions like `set_gpu_limit` and `add_node_selector_constraint` to facilitate GPU resource configuration within pipeline components. The user's contributions extended to adding support for TPU settings in the DSL by implementing add_pod_annotation and adding `use_tpu` utility.
pipelinetektondata-sciencemachine-learningmlops
hongye-sun/pipelines
Dec 2018 - Sep 2020
Machine Learning Pipelines for Kubeflow
Contributions:196 pushes, 56 branches in 1 year 9 months
data-sciencemachine-learningmlopskubeflowkubernetes
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.