Chen Liang
Staff Research Scientist at Google DeepMind
Mountain View, California, United States
Summary
🤩
Rockstar🎓
Top SchoolChen Liang is a Staff Research Scientist at Google DeepMind with 11 years of experience building and deploying cutting-edge AI systems across deep learning, reinforcement learning, natural language understanding, program synthesis and AutoML. He led the AutoML project that discovered the Lion optimizer—now used in production across vision, multimodal, image generation and large language models—and has driven significant revenue and cost improvements for Google products. Chen blends research and engineering: he has open-sourced AutoML-Zero components, integrated Lion into TensorFlow/Optax, and contributed low-level dataset and C++ utilities for Google Research. He holds a PhD in AI and Cognitive Science from Northwestern and is known for mentoring teams and publishing at top venues (NeurIPS, ICML, ICLR). A less obvious strength is his knack for turning program-search discoveries into widely adopted production optimizers and architectures, bridging novel research with real-world impact.
10 years of coding experience
9 years of employment as a software developer
Bachelor's Degree Physics, Bachelor's Degree Physics at Peking UniversityDoctor of Philosophy (Ph.D.) Artificial Intelligence and Cognititve Science, Doctor of Philosophy (Ph.D.) Artificial Intelligence and Cognititve Science at Northwestern UniversityZhenResidence
Chinese, English
Github Skills (14)
algorithm10
machine-learning10
c-language10
tensorflow10
cprogramming-language10
automl10
opt10
python10
optimisation10
optimizers10
optim10
optimization10
github9
ai9
Programming languages (3)
Github contributions (5)
google-research/google-research
Mar 2020 - Apr 2021
Google Research
Role in this project:
ML Engineer
ML EngineerContributions:26 commits, 6 comments in 1 year 2 months
Contributions summary:Chen contributed to the AutoML-Zero codebase, specifically open-sourcing the AutoML-Zero code base. The user worked on dataset utilities, introducing and modifying code related to dataset generation, downcasting, and dataset creation using various regression creators. The user's changes included modifications to header files, demonstrating familiarity with C++ and the project's underlying machine learning concepts.
googlemachine-learningai
google/automl
Feb 2023 - Mar 2025
Google Brain AutoML
Role in this project:ML Engineer
ML EngineerContributions:3 PRs, 21 pushes, 2 comments in 2 years
Contributions summary:Chen implemented and integrated the Lion optimizer, a novel optimization algorithm, within the TensorFlow and Optax frameworks. Their contributions focused on adding the Lion optimizer implementation, fixing code typos, and improving documentation, including linking relevant research papers and adding citations. The user's work involved modifying core optimization routines and integrating them into the existing AutoML codebase, demonstrating proficiency in model optimization techniques.
brainefficientnetv2pythonefficientnetdeep-learning
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.