Hi, I'm a PhD student at USC, advised by Professor Seo Jin Park. I have a broad interest in distributed and operating systems. I also build systems for more efficient LLM inference.
University of Southern California
Ph.D. in Computer Science
September 2025 - Present
New York University
B.A. in Computer Science with honors, Minor in Math
January 2021 - May 2024
GPA: 3.866/4.0
Related CS Courses (A-Grades): Operating Systems, Database Systems, Algorithms, Computation Theory, Formal Verification
Math Courses: Honors series in Vector Calculus, Analysis, and Abstract Algebra
Research Intern at ByteDance
May 2026 - August 2026
- Optimizing batch inference system.
Full-time Research Intern at USC Networked Systems Lab
July 2024 - May 2025
- High-resolution, low-overhead application profiling and resource demand heterogeneity.
- Preliminary research on massive-parallel, burst-compatible cloud infrastructure.
Undergraduate Researcher at NYU Secure Systems Lab
June 2023 - June 2024
- Efficient and secure software fault isolation for the cloud. Contributions made IPC benchmarks 75-200% faster than native Linux.
- Adoption of WASM runtimes and porting glibc to WASM.
- Democratizing MoE LLM Decoding via Barrier-Free Expert Parallelism
Yizhuo Liang*, Shaoyu Wang*, Jaeyong Song, Yanqi Zhou, Geon-Woo Kim, Guangrong He, Seo Jin Park
The 32nd ACM Symposium on Operating Systems Principles (SOSP '26), Oct 2026
* Equal contribution - Moebius: Serving Mixture-of-Expert Models with Seamless Runtime Parallelism Switch
Shaoyu Wang, Yizhuo Liang, Jaeyong Song, Chong Li, Seo Jin Park
arXiv:2606.26607 (Preprint), Jun 2026 - Cornserve: Efficiently Serving Any-to-Any Multimodal Models
Jeff J. Ma, Jae-Won Chung, Jisang Ahn, Yizhuo Liang, Akshay Jajoo, Myungjin Lee, Mosharaf Chowdhury
arXiv:2512.14098 (Preprint), Dec 2025 - Granular Resource Demand Heterogeneity
Yizhuo Liang, Ramesh Govindan, Seo Jin Park
The 20th Workshop on Hot Topics in Operating Systems (HotOS '25), May 2025
© 2025 Yizhuo Liang
Welcome to fork my theme!