Summary
Chung-hsuan Tung is a Ph.D. candidate and graduate research assistant at Duke University with eight years of experience at the intersection of wireless baseband processing, computer architecture, and ML hardware reliability. He specializes in real-time software implementations for mmWave (5G FR2) baseband processing—optimizing MIMO matrix operations and delivering DSP C++/AVX-512 solutions that achieve high throughput on a single CPU core. His prior work on ReRAM-based CNN accelerators focused on built-in self-test, fault characterization, and dynamic task remapping to tolerate stuck-at-faults, bridging device-level failures to system-level robustness. Internships at NVIDIA and AMD reinforced his tooling and fault-modeling expertise across safety, resilience, and bare-metal AI Engine development. Known for turning cross-layer research into practical edge-server and vRAN testbed solutions, he combines rigorous academic performance (PhD, 3.97/4) with hands-on system optimization.
8 years of coding experience
1 year of employment as a software developer
Bachelor of Science - BS, Electrical Engineering, 4.22/4.3, Bachelor of Science - BS, Electrical Engineering, 4.22/4.3 at National Cheng Kung University
Doctor of Philosophy - PhD, Electrical and Computer Engineering, 3.97/4, Doctor of Philosophy - PhD, Electrical and Computer Engineering, 3.97/4 at Duke University
High School Diploma, High School Diploma at Taipei Municipal Jianguo High School
Chinese, English, Japanese, German