Zhi Lin
Machine Learning Engineer at Intel
Songjiang District, Shanghai, China
Summary
👤
Senior🎓
Top SchoolZhi Lin is a Machine Learning Engineer with 8 years' experience building high-performance ML and big-data systems, currently at Intel in Shanghai. He brings a strong systems background from internships at AWS and Intel—optimizing CUDA kernels, implementing sparse matrix partitioning and JIT compilation for GNN workloads, and integrating DAOS with Apache Spark via a thread-safe Java JNI layer. An active open-source contributor to the popular Ray project, he added Java object-store APIs, dataset Spark interoperability, and robustness features that improve distributed data handling. His work sits at the intersection of ML kernels, distributed runtime engineering, and DevOps-driven CI, enabling significant runtime speedups and scalable data pipelines. Zhi combines academic training from USC and SJTU with hands-on low-level optimization skills, often surfacing engineering trade-offs that reduce preprocessing costs by orders of magnitude.
8 years of coding experience
1 year of employment as a software developer
Master of Science - MS, Computer Science, Master of Science - MS, Computer Science at University of Southern CaliforniaBachelor of Science - BS, Electrical and Computer Engineering, 3.6/4.0, Bachelor of Science - BS, Electrical and Computer Engineering, 3.6/4.0 at UM-SJTU Joint Institute, Shanghai Jiao Tong University
Github Skills (11)
javas10
ray10
distributed-systems10
spark10
python10
java10
data-science9
data-engineering9
deep-learning8
machine-learning8
deploying7
Programming languages (4)
Github contributions (5)
ray-project/ray
Jul 2021 - Sep 2022
Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.
Role in this project:
Back-end Developer & DevOps Engineer
Back-end Developer & DevOps EngineerContributions:40 reviews, 6 commits, 24 PRs in 1 year 2 months
Contributions summary:Zhi contributed to the Ray project by implementing Java APIs for object store management, specifically focusing on assigning object ownership within the `Ray.put()` function. The user added functionality related to tracing, ensuring it is not enabled when tracing is not necessary. They also made significant changes to the dataset component, implementing `from_spark`, `to_spark` functions, and other optimizations to improve data handling. Furthermore, the user contributed to the Java API and added maxTaskRetries option for java actor creation, and re-enabled RayDP dataset tests.
pythonconsistsruntimetensorflowserving
kira-lin/ray
Apr 2021 - May 2024
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:136 pushes, 27 branches in 3 years 1 month
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.