Hanmei Yang
Research Scientist at Meta
Menlo Park, California, United States
Summary
🤩
Rockstar🎓
Top SchoolHanmei Yang is a research scientist at Meta based in Menlo Park with eight years of experience specializing in efficient ML systems and memory- and performance-optimization for training and inference. Her PhD work at UMass Amherst focused on NUMA-aware memory management, multi-convolution execution, and optimizing large-language-model workflows on resource-limited devices, bridging deep systems research with practical deployment. She has iterated through roles at Meta from intern to visiting researcher to research scientist, and earlier internships at ByteDance where she implemented and benchmarked mobile inference allocation strategies. Hanmei combines low-level systems skills (CUTLASS, TensorRT, Rammer) with applied engineering experience building low-latency trading infrastructure and mobile apps, showing comfort across research prototypes and production concerns. Notably, she has repeatedly targeted memory allocation and tensor lifecycle problems that are often overlooked but crucial for cost- and latency-constrained ML deployments. She brings a pragmatic researcher’s mindset focused on measurable speedups and deployable system improvements.
8 years of coding experience
4 years of employment as a software developer
Bachelor's degree, Computer Science and Technology, Bachelor's degree, Computer Science and Technology at 南方科技大学Visiting Student, Computer Science, Visiting Student, Computer Science at 美国加州大学尔湾分校Summer Program, Summer Program at The University of British ColumbiaDoctor of Philosophy - PhD, Electrical and Computer Engineering, Doctor of Philosophy - PhD, Electrical and Computer Engineering at University of Massachusetts Amherst
Chinese, English
Github Skills (57)
pytorch10
python10
acceleration10
machine-learning10
optimizer10
inference10
numpy10
deep-learning10
tensorflow10
gpu10
autograd10
neural-network10
offloading10
cuda10
quantization10
Programming languages (13)
Github contributions (5)
NicoleMayer/ECE668-Notes
Aug 2020 - Dec 2020
Here are notes for ECE668 Computer Architecture in UMass, taught by Yadi Eslami
Contributions:1 release, 46 commits, 42 pushes in 3 months
notesarchitecturecomputer-architecture
NicoleMayer/MassLab-HomePage
Jan 2021 - Jul 2022
Home page for Mass Lab
Contributions:22 commits, 4 PRs, 25 pushes in 1 year 6 months
home-pagelabmass
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.