Cheng-hsiang Chiu is a systems and performance-focused software engineer with five years of experience designing task-graph runtimes, scheduling algorithms, and heterogeneous CPU/GPU execution models. He has contributed to Intel’s SYCL Graph runtime and production EDA performance at Cadence, and his work on Taskflow test automation underpins robust pipeline correctness for a widely used C++ task-parallel library. His background spans large-scale monitoring at CERN, energy-efficient edge inference for Arctic wildlife, and parallelized scientific workflows, reflecting a rare mix of low-level optimization and applied research. Recently he’s extended into AI systems and LLM infrastructure, building end-to-end apps with OpenAI, embeddings, PyTorch, and vector DBs while maintaining a profiling-driven, correctness-first approach. Based in Madison, Wisconsin, he combines PhD-level research training with hands-on engineering to push systems toward higher concurrency and real-world scale.
5 years of coding experience
7 years of employment as a software developer
Doctor of Philosophy - PhD Electrical and Computer Engineering, Doctor of Philosophy - PhD Electrical and Computer Engineering at University of Wisconsin-Madison
The University of Utah
Master's degree Communication Engineering, Master's degree Communication Engineering at National Chiao Tung University
Bachelor's degree Electrical Engineering, Bachelor's degree Electrical Engineering at National Chung Cheng University
Master's degree Computer Science, Master's degree Computer Science at EPFL
A General-purpose Task-parallel Programming System using Modern C++
Role in this project:
Back-end Developer & Test Automation Engineer
Contributions:8 commits, 26 pushes, 4 branches in 1 year
Contributions summary:Cheng-hsiang primarily contributed to the testing framework for the taskflow library, focusing on unit tests for pipeline functionalities. They implemented and updated test cases, covering different pipeline configurations, including serial and (commented-out) parallel pipe types. The user's work involved verifying the correctness of data flow and token management within the pipelines, using the doctest framework to validate expected behaviors and performance.
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.