My name is Chengjhih Shih. I’m a first-year PhD student at Georgia Tech’s EIC Lab, supervised by Prof. Yingyan (Celine) Lin. My research interests include computer architecture and high-performance computing, with a focus on 3D neural rendering acceleration and chiplet accelerators.

Will be more descriptions about my PhD journey in the future

Before coming to Georgia Tech, I received my BS and MS from the Department of Computer Science at National Taiwan University (NTU), where I worked on a variety of research projects:

Accelerating Quantum Simulations

I worked with Prof. Shih-Hao Hung at NTU on quantum annealing acceleration, restructuring simulated quantum annealing (SQA) using CUDA and Tensor Cores, achieving an 86X speedup. This was presented at ISC High Performance 2022. I also contributed to large-scale circuit-based quantum simulations on NVMe disks, showcased at Future Tech in Taiwan.

Software Optimization and Hardware Design for Cryptosystems

In collaboration with Prof. Bo-Yin Yang at Academia Sinica, I accelerated polynomial multiplication in lattice-based cryptosystems using the NTT, with ARM assembly optimizations for Saber and NTRU, accepted at Cryptographic Hardware and Embedded Systems (CHES). I also designed an FPGA for the multivariate cryptosystem “oil and vinegar,” submitted to NIST PQC Digital Signature Schemes.

Specialization and Co-Design for Homomorphic Encryption (HE)

Since 2022, I’ve worked with Wistron on HE hardware acceleration. My focus included FPGA synthesis and multi-FPGA prototypes, with work showcased at Mobile Internet Security 2023, forming the basis of my master’s thesis.

Contact Info

• Email: cs861324@gmail.com
• 266 Ferst Dr NW, Room 3206, Atlanta, GA 30332