Carson Wang
AI Framework Senior Staff Engineer at Intel Corporation
Shanghai, Shanghai, China
Summary
🤩
Rockstar🎓
Top SchoolCarson Wang is an AI framework senior staff engineer based in Shanghai with 11 years of experience building high-performance distributed AI and data systems. At Intel he has driven production-grade enablement of LLM serving across CPUs, GPUs, and Gaudi accelerators, open-sourcing platforms like LLM-on-Ray and RayDP to unify Data+AI pipelines. His background includes deep kernel-to-application work—optimizing Spark, Adaptive Execution, Apache Gluten and Alluxio for exascale and hardware acceleration—plus hands-on bug fixes in prominent open-source projects such as Apache Spark and HiBench. He combines manager-level leadership with hands-on engineering, having led performance tuning for torch.compile-based inference stacks (vLLM) while still contributing low-level fixes to storage and streaming components. Not obvious from his title: he started in developer support and mobile-game platform engineering, giving him a rare operational empathy for debugging complex production issues. He holds a Master’s in Software Engineering from Fudan and consistently bridges research-grade optimizations with pragmatic, deployable systems.
11 years of coding experience
14 years of employment as a software developer
Master's degree Software Engineering, Master's degree Software Engineering at Fudan UniversityBachelor's degree Information Engineering, Bachelor's degree Information Engineering at East China University of Science and Technology
Github Skills (18)
filesystem10
spark10
back-end-development10
testing10
java10
scala10
webui10
data-storage10
javas10
orchestrate10
data-streaming10
orchestration10
kafka9
yarnpkg9
yarn39
Programming languages (6)
Github contributions (5)
Intel-bigdata/HiBench
Nov 2015 - Nov 2022
HiBench is a big data benchmark suite.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:3 releases, 17 reviews, 196 commits in 7 years 1 month
Contributions summary:Carson primarily focused on fixing bugs related to data generation and streaming in the HiBench benchmark suite. Their commits involved modifying the `NewKafkaConnector` and related Java files, specifically addressing issues in data publishing and file offsets. The contributions demonstrate a focus on improving the accuracy and reliability of the streaming data generation process within the context of the HiBench big data benchmark suite. They also corrected a file offset issue in the source code.
benchmark-databenchmarkingbenchmark-suitebig-databenchmark
Alluxio/alluxio
Jan 2015 - Jan 2016
Alluxio, data orchestration for analytics and machine learning in the cloud
Role in this project:Back-end Developer
Back-end DeveloperContributions:106 commits, 39 PRs, 137 comments in 11 months
Contributions summary:Carson primarily contributed to the backend of the Alluxio project, focusing on data orchestration. Their commits addressed critical issues in the TachyonFile class, such as handling block index out-of-bounds exceptions and refactoring the logic for getting location hosts. They also worked on core components, including fixing Jetty-related issues in the web UI, and optimizing remote reading from the hierarchy store. Additionally, the user implemented improvements in handling non-existent paths within the TFsShell.
data-analysisvirtual-distributed-filesystemanalyticsorchestrationdata-science
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.