Guangxiao H
Research Scientist at Alarm.com
Vienna, Virginia, United States
Summary
👤
Senior🎓
Top SchoolGuangxiao H is a research scientist with 8 years of experience applying deep learning to video understanding, object detection, and multi-agent systems. Based in Vienna, VA, he develops vision foundation models and end-to-end pipelines that combine RLHF-tuned agents, automatic data/ground-truth generation, and high-precision auto-labeling for production analytics at Alarm.com. Previously at ObjectVideo Labs he optimized neural architectures for view-change detection and re-identification under challenging appearance and illumination shifts, and as an IBM postdoc he designed multi-stream activity-recognition networks. His work sits at the intersection of applied research and production engineering, routinely translating experimental models into scalable pipelines with business-oriented metrics. Notably, he blends reinforcement learning for agent behavior with practical computer-vision solutions to reduce manual labeling and accelerate model deployment.
8 years of coding experience
3 years of employment as a software developer
The University of Maryland, College ParkBachelor of Science (B.S.), Electrical Engineering, Bachelor of Science (B.S.), Electrical Engineering at Peking University
Chinese, English
Github Skills (6)
docker5
nvidia-docker5
cuda5
nvidia4
gpu4
docker-run2
Github contributions (2)
GigiCoding/video-summarization
Jul 2017 - May 2018
Contributions:20 commits, 17 pushes, 1 branch in 10 months
NVIDIA/nvidia-docker
Jun 2019 - Jun 2019
Build and run Docker containers leveraging NVIDIA GPUs
Contributions:1 comment in 1 day
cudanvidia-dockercontainersnvidialeveraging
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.