Huansong Fu is a Staff Software Engineer with eight years of experience building and optimizing distributed systems, high-performance computing runtimes, and large-scale database engines. He blends deep research credentials (PhD work and postdoctoral work at Argonne) with production impact at companies like Broadcom and AWS, where he led Greenplum kernel evolution and delivered a 10x columnar storage speedup and read-replica features. Skilled in MPI, OpenSHMEM, RDMA, Spark, and JVM/C++ ecosystems, he contributes regularly to the widely used MPICH repository, improving RMA performance and correctness in the OFI network layer. Comfortable toggling between low-level HPC optimizations and customer-facing production fixes, he has a track record of resolving P0 incidents and improving runtime performance by 30% in research projects. Based in the New York City area, he pairs systems-thinking with hands-on coding and cluster management experience, making him effective at turning academic advances into scalable product features.
8 years of coding experience
7 years of employment as a software developer
Doctor of Philosophy (PhD), Computer Science, Doctor of Philosophy (PhD), Computer Science at Florida State University
Bachelor of Science (BS), Computer Science, 3.7/4.0 (2 of 143), Bachelor of Science (BS), Computer Science, 3.7/4.0 (2 of 143) at University of Electronic Science and Technology
Master of Science (MS), Computer Science, Master of Science (MS), Computer Science at Auburn University
Contributions:32 commits, 12 PRs, 3 pushes in 1 year 1 month
Contributions summary:Huansong primarily contributed to the MPICH repository by modifying the code related to Remote Memory Access (RMA) operations, specifically within the OFI (Open Fabrics Interface) network module. Their work focused on improving code consistency, optimizing performance, and fixing potential bugs related to data size checks and the handling of MPI_PROC_NULL. The user's contributions involved refactoring the code, removing redundant checks, and adding specific handling for contiguous buffers to enhance the efficiency of RMA operations.
Greenplum Database - Massively Parallel PostgreSQL for Analytics. An open-source massively parallel data platform for analytics, machine learning and AI.
Contributions:751 pushes, 340 branches, 1 comment in 2 years 5 months
analyticsparallelsqlgreenplummachine-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.