I am a Ph.D. student in the
School of Computer Science at
Peking University, co-advised
by Prof. Xuanzhe Liu and Prof. Xin Jin.
I received my B.S. degree (Summa Cum Laude) in artificial intelligence from Turing Class, Peking
University, in 2022.
My research interests include distributed systems, resource disaggregation, and their intersection with
serverless computing and LLM serving.
Interests
- Serverless Computing
- Distributed Systems
- Resource Disaggregation
Education
-
Ph.D. in Computer Science
2022 - present
Peking University -
B.S. in Artificial Intelligence
Summa Cum Laude
2018 - 2022
Peking University
Selected Publications
KUBEDIRECT: Unleashing the Full Power of the Cluster Manager for Serverless Computing
Sheng Qi, Zhiquan Zhang, Xuanzhe Liu, Xin Jin
USENIX NSDI, 2026
[PDF]
[DOI]
HydraServe: Minimizing Cold Start Latency for Serverless LLM Serving in Public Clouds
Chiheng Lou, Sheng Qi, Chao Jin, Dapeng Nie, Haoran Yang, Xuanzhe Liu, Xin Jin
USENIX NSDI, 2026
[PDF]
[DOI]
Efficient Fault Tolerance for Stateful Serverless Computing with Asymmetric Logging
Sheng Qi, Haoyu Feng, Xuanzhe Liu, Xin Jin
ACM TOCS, 2025
[PDF]
[DOI]
Halfmoon: Log-Optimal Fault-tolerant Stateful Serverless Computing
Sheng Qi, Xuanzhe Liu, Xin Jin
ACM SOSP, 2023
[PDF | Extended]
[Slides]
[Code]
[DOI]
Experience
- Research Intern, WXG, Tencent
2025.5 - 2025.8- Use Context Parallelism to optimize long sequence prefill
- 2x speedup vs Tensor Parallelism and near-linear scaling with up to 32 GPUs
- Research Intern, PAI, Alibaba Cloud
2026.2 - present- Load balancing across Qwen family models