Hanming Lu
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolHanming Lu is an AI research scientist with five years of experience specializing in inference and LLM performance engineering across startups and large tech, currently working on inference at Meta. He previously contributed to xAI as Member of Technical Staff and built CUDA kernels and performance optimizations for large language models at Anyscale, bridging research and production needs. Trained at UC Berkeley (MS/PhD) with foundational work from the University of Waterloo, he combines rigorous academic research with hands-on systems engineering. Hanming’s strengths are squeezing latency and cost out of model inference pipelines and translating novel model ideas into deployable, high-performance code. Colleagues appreciate that he moves comfortably between low-level GPU work and higher-level model evaluation, making him effective at both prototyping and production hardening. Based in San Francisco, he brings a practical researcher’s mindset to real-world ML systems challenges.
5 years of coding experience
3 years of employment as a software developer
Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at University of California, BerkeleyBachelor’s Degree, Computer Science, Bachelor’s Degree, Computer Science at University of Waterloo
Github Skills (42)
pytorch10
parallel10
rllib10
python10
data-science10
machine-learning10
inference10
large-language-models10
java10
scaling10
reinforcement-learning10
llm-inference10
llm10
hyperparameter-optimization10
mlops10
Programming languages (4)
Github contributions (5)
hanming-lu/dc-replication
Dec 2021 - Oct 2022
DataCapsule Replication for Paranoid Stateful Lambda (PSL) / Global Data Plane (GDP)
Contributions:59 commits, 22 PRs, 39 pushes in 10 months
pslgdpparanoidplanestateful
hanming-lu/ray
Oct 2020 - Jun 2022
An open source framework that provides a simple, universal API for building distributed applications. Ray is packaged with RLlib, a scalable reinforcement learning library, and Tune, a scalable hyperparameter tuning library.
Contributions:18 pushes, 5 branches in 1 year 7 months
apirayscalabledistributed-applicationshyperparameter
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.