Kuan-yu Chen
Software Engineer at SiFive
Tainan Metropolitan Area Taiwan
Summary
👤
Senior🎓
Top SchoolKuan-yu Chen is a software engineer with a decade of experience specializing in high-performance neural network inference and embedded AI. Currently pursuing a master's in AI at National Cheng Kung University, he blends deep learning research—especially knowledge distillation, model compression, and contrastive learning—with hands-on systems work from his Raspberry Pi autonomous vehicle project to production-grade kernels. At SiFive he implemented and optimized integer matrix kernels and vectorized NN operators on RISC-V vector extensions, achieving multi-fold speedups through instruction-level tuning and memory-layout optimizations. He bridges research and engineering by integrating low-level optimized kernels into frameworks like IREE and building internal profiling tools to continually improve throughput. Notably, he combines practical edge-device experimentation with RTL/FPGA-informed performance engineering, enabling smaller models to retain strong accuracy on constrained hardware.
10 years of coding experience
學士, Information Technology, 學士, Information Technology at 中原大學碩士, 人工智慧科技碩士學位學程, 碩士, 人工智慧科技碩士學位學程 at 國立成功大學
Github Skills (70)
risc10
jtag10
capstone10
risc-v10
glibc9
unicorn9
utm9
x86-649
yocto9
security9
vms9
arm9
assembly9
defold9
luks9
Programming languages (5)
Github contributions (5)
nylon7/Igh-Ethercat-master
Aug 2015 - Aug 2015
Contributions:25 commits, 22 pushes, 1 branch in 1 day
nylon7/opensbi
Aug 2019 - Aug 2019
RISC-V Open Source Supervisor Binary Interface
Contributions:13 pushes, 1 branch in 17 days
risc-vriscvriscsupervisor
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.