Hongze Zhang is a software engineer with 11 years' experience specializing in data infrastructure, query optimizers and accelerator projects, currently working on Apache Gluten at IBM to offload JVM-based SQL execution to native engines. He has a strong open-source footprint across Apache Arrow, Calcite, and Velox, contributing cross-language (C++/Java) integrations, memory management improvements, and spillable join implementations that boost analytics performance. At Intel and prior roles at Tencent and Alibaba he focused on high-performance analytics stacks and universal SQL engines on large-scale storage, bringing production-grade optimizations to cutting-edge compute architectures. Notably, he implemented Arrow C ABI-based data interchange and robust JNI handling in Gluten, demonstrating deep systems-level expertise that bridges runtimes. Based in Dublin, he combines backend systems craftsmanship with pragmatic leadership from early team-lead experience, grounded in a B.E. in Microelectronics.
11 years of coding experience
10 years of employment as a software developer
Bachelor of Engineering (B.E.), Microelectronics, Bachelor of Engineering (B.E.), Microelectronics at Soochow University (CN)
Gluten is a middle layer responsible for offloading JVM-based SQL engines' execution to native engines.
Role in this project:
Back-end Developer
Contributions:2058 reviews, 81 commits, 1356 PRs in 8 months
Contributions summary:Hongze primarily contributed to the Gluten project by modifying and enhancing the codebase related to Apache Arrow integration. The commits focused on integrating main stream code of Apache Arrow, reworking the JNI loading process, and utilizing the Arrow C ABI for data interchange. The user also worked on implementing data interchange and record batch handling between C++ and Java, and implemented exception handling within the C++ code.
A composable and fully extensible C++ execution engine library for data management systems.
Role in this project:
Back-end Developer
Contributions:182 reviews, 64 PRs, 4 branches in 4 years 8 months
Contributions summary:Hongze primarily focused on implementing features related to memory management and query execution within the Velox framework. They added support for threshold-based hash join build spilling, including implementing the option and integrating it into the testing framework. The user also introduced the ability to create a memory manager with a custom arbitrator and fixed issues related to static initialization and single-threaded execution mode. They also addressed a like function error.
cppdata-managementquery-processing
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.