Zhenyuan Yang Operating Systems & ML Systems

Operating Systems & ML Systems

Zhenyuan Yang

Master’s student at the Institute of Software, Chinese Academy of Sciences

My research spans operating systems and machine learning systems, with a focus on efficient and reliable inference across GPUs, resource-constrained devices, and collaborative edge–cloud systems. This site brings together selected research, open-source projects, and practical notes from systems work.

Current Research Intern · Huawei 2012 Laboratories

Explore research Read notes Email me
GitHub ORCID
Portrait of Zhenyuan Yang

Research & engineering

Selected Work

Publications

3

Peer-reviewed work, preprints, and ongoing research.

  • Harmonizing PIM-Accelerated LLM Inference and Mobile Applications via Spatial-Temporal Isolation
    Zhenyuan Yang, Shicheng Deng, Zhao Zhang, Mingyu Li, Leibin Ni, Han Lin, Mengying Guo, Tao Wang, Haibo Chen
    Under review 2026
  • Determinism-Preserving GPU Spatial Sharing with Vitamin-E
    Zhenyuan Yang, Wenxin Zheng, Mingyu Li, Haibo Chen
    Under ReviewarXiv:2603.15042 2026
  • Towards Large Language Model-Friendly APIs
    Yuan Wang, Zhenyuan Yang, Zhanbo Wang, Mingyu Li, Zhilin Wu, Haibo Chen
    PublishedACM SIGOPS Operating Systems Review59(1):17–23 2025
View all research

Awards

3

Selected academic honors and research awards.

  • First Place, ACM Student Research Competition at SOSP
    ACM SRC Grand Finals 2026 Testimonial Paper (PDF)
  • Outstanding Graduate, Nanjing University
  • Outstanding Student, Nanjing University

Projects

2

Open-source software and selected engineering work.

  • SoulKnight A roguelike game built with LibGDX and Netty.
  • vGPU A library for intercepting and dispatching GPU kernel launches.
View on GitHub

Notes & essays

Recent Writing

Browse all