Mingliang Zhu
Hangzhou City, Zhejiang, China
Summary
🤩
RockstarMing Zhu is a principal engineer and systems architect with 8 years of industry experience building large-scale ML, data and inference infrastructure across Microsoft, Facebook, Google, and now Shopify. He pairs deep systems and networking roots (PhD work on content-centric data center networks and early OpenFlow contributions) with hands-on backend engineering in C++/Python, optimizing query engines and ML pipelines. His open-source contributions to high-profile projects like Apache Spark and Kyuubi show expertise in SQL optimization, Delta Lake integration, and performance-sensitive features such as join/winow rule tuning and null handling. At Microsoft and Google he led deep learning inference and AI-on-GKE efforts for high-throughput services, and at Shopify he architects ML/data platforms for production workloads. Colleagues rely on him for bridging research-grade designs with pragmatic, production-ready implementations—he’s equally comfortable rewriting worker managers as he is extending word2vec and image-processing pipelines.
8 years of coding experience
Github Skills (20)
spark-sql10
spark10
delta-lake10
word2vec10
back-end-development10
image-processing10
big-data10
machine-learning10
java10
scala10
javas10
sql10
backend10
opencv9
data-structures8
Programming languages (5)
Github contributions (5)
apache/spark
Jan 2021 - Feb 2026
Apache Spark - A unified analytics engine for large-scale data processing
Role in this project:
Back-end Developer
Back-end DeveloperContributions:110 reviews, 35 PRs, 189 comments in 5 years 1 month
Contributions summary:Mingliang contributed significantly to the Apache Spark codebase by addressing multiple SQL-related issues. Their work involved optimizing query execution, specifically focusing on the `RemoveRedundantAggregates` and `OptimizeJoinCondition` rules. The commits demonstrate a deep understanding of Spark's internal optimization processes and how to improve query performance through code modifications. The user also implemented changes related to window functions and handling null values in join conditions.
apache-sparkpythonscalarjava
byzer-org/byzer-lang
Jun 2018 - May 2021
Byzer (former MLSQL): A low-code open-source programming language for data pipeline, analytics and AI.
Role in this project:Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:47 commits, 32 PRs, 10 comments in 3 years
Contributions summary:Mingliang primarily contributed to image processing functionalities within the Byzer-lang project, including the addition of image resizing capabilities using Java and OpenCV. They integrated and tested image processing logic using Scala, Spark, and Java libraries like Imgscalr. Furthermore, the user extended the word2vec model to add merge parameters. This indicates a focus on data processing, machine learning, and potentially performance enhancements.
aidata-pipelinelow-codeprogramming-languagemachine-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.