Hongwu Peng
Research Scientist Engineer II (Senior) at Adobe
Seattle, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolHongwu Peng is a research scientist engineer with seven years of experience at the intersection of generative models, privacy-preserving machine learning, and ML systems, currently co-leading foundation model training and modeling efforts at Adobe in Seattle. His work spans academia and industry, with publications at NeurIPS, ICCV, MICRO and other top conferences and a 2022 Taylor L. Booth predoctoral fellowship recognizing him as UConn’s top Ph.D. student. He has driven model-scaling and efficiency projects for >100B-parameter models, Mixture-of-Experts architectures, and GenAI text-to-image/video systems during internships and research roles at Adobe, Microsoft, and national labs. Comfortable across GPU/FPGA systems and large-scale LLM training pipelines, he combines systems-level optimization with algorithmic research in privacy and generative modeling. Hongwu’s background in electrical and computer engineering informs a hardware-aware approach to ML, and his Google Scholar and personal site reflect a steady output of high-impact, applied research.
7 years of coding experience
5 years of employment as a software developer
Electrical and Electronics Engineering, Electrical and Electronics Engineering at Huazhong University of Science and TechnologyMaster’s Degree Electrical Engineering, Master’s Degree Electrical Engineering at University of ArkansasDoctorate Degree Computer Engineering, Doctorate Degree Computer Engineering at Clemson University鄂南高级中学Doctorate Degree Computer Science and Engineer, Doctorate Degree Computer Science and Engineer at University of Connecticut
Github Skills (37)
pytorch10
anchor10
python10
data-science10
temporal10
architecture10
imagenet10
lifecycle10
incremental-learning10
machine-learning10
parameter-tuning10
machine-learning-algorithms10
feature-engineering10
model-compression10
hyper10
Programming languages (2)
Github contributions (5)
microsoft/Cream
Oct 2020 - Aug 2022
This is a collection of our NAS and Vision Transformer work.
Contributions:2 releases, 67 commits, 25 PRs in 1 year 10 months
pytorchimagenetvision-transformerknowledge-distillationvision
researchmm/TracKit
Jun 2020 - Oct 2021
[ECCV'20] Ocean: Object-aware Anchor-Free Tracking
Contributions:28 commits, 25 pushes, 11 comments in 1 year 4 months
oceaneccvtracking-freeobject-detectioncomputer-vision
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.