Wenhao Wu
Bellevue, Washington, United States
Summary
👤
SeniorWenhao W is an Applied Scientist II at Amazon AGI specializing in generative foundation models for video post-training and multimodal reasoning, with a PhD in Computer Science from the University of Sydney and nine years of research and industry experience. He has a strong publication track record across top-tier venues (CVPR, ICCV, AAAI) and led work on large-scale video-language pretraining, self-supervised learning, and semantic video understanding at Baidu VIS under senior mentorship. His recent focus bridges video-capable LLMs and chain-of-thought techniques for zero-shot vision tasks, reflecting a rare mix of deep vision expertise and practical multimodal model engineering. Wenhao has contributed to award-winning applied research (traffic anomaly detection, deblurring challenges) and moved between research labs and production teams at companies including Amazon, Baidu, Snap, SenseTime, and Samsung. Based in Bellevue, WA, he combines academic rigor with product impact—often turning sophisticated theory into scalable model components for real-world video understanding. Notably, his trajectory shows a pattern of short, high-impact industry collaborations that accelerate prototype-to-production cycles in multimodal AI.
9 years of coding experience
Github Skills (47)
re-identification10
classifier10
video-understanding10
action-recognition10
transfer-learning10
visual-recognition10
temporal9
benchmark9
x3d9
openmmlab9
ava9
efficientnet9
resnet9
fusion9
language-model8
Programming languages (2)
Github contributions (5)
whwu95/whwu95.github.io
Jun 2020 - Mar 2023
Contributions:88 pushes, 1 branch in 2 years 9 months
whwu95/BIKE
Jan 2023 - Sep 2024
【CVPR'2023】Bidirectional Cross-Modal Knowledge Exploration for Video Recognition with Pre-trained Vision-Language Models
Contributions:1 release, 2 PRs, 18 pushes in 1 year 8 months
language-modelaction-recognitioncross-modal-learningvideo-language-understandingvideo-recognition
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.