University of Edinburgh 爱丁堡大学
Yanwei Ye is a Ph.D. student in Computer Science at the University of Edinburgh, working on systems for next-generation AI accelerators and agentic AI. His research sits at the intersection of AI systems, computer architecture, compilers, and runtime infrastructure. He studies how to make emerging AI workloads efficient on heterogeneous and wafer-scale hardware, including low-latency scheduling for agentic applications, compilation and execution planning for wafer-scale accelerators, and performance modeling for model-serving systems. More broadly, he is interested in turning advances in AI models and accelerator hardware into reliable, programmable, and scalable systems. He previously completed an M.S. in Computer Science at the University of Science and Technology of China.
叶言威是爱丁堡大学计算机科学博士生，研究方向聚焦于面向下一代 AI 加速器的系统软件与智能体 AI。其研究结合 AI 系统、计算机体系结构、编译器和运行时，关注如何让新型 AI 工作负载在异构计算平台和晶圆级 AI 加速器上高效运行。当前主要研究包括智能体应用的低延迟调度、面向晶圆级 AI 加速器的编译与执行计划，以及模型服务系统的性能建模。更广泛地，他希望连接 AI 模型、加速器硬件与系统软件，构建高效、可靠、可编程且可扩展的 AI 基础设施。此前，他在中国科学技术大学获得计算机科学硕士学位。
Education教育经历
-
University of Edinburgh爱丁堡大学Ph.D. in Computer Science计算机科学博士Sep. 2025 – Jun. 20292025年9月 – 2029年6月
-
University of Science and Technology of China中国科学技术大学M.S. in Computer Science计算机科学硕士Sep. 2022 – Jun. 20252022年9月 – 2025年6月
Experience经历
-
Tencent腾讯Guangzhou广州
Selected Publications代表性论文 (view all查看全部 )
SwarmPilot: Agentic Scheduling for Agentic Systems
Yeqi Huang*, Yanwei Ye*, Guomin Chen, Wenhao Su, Bin Gong, Jialian Li, Zhan Lu, Yangshen Deng, Xuan Sun, Le Xu, Luo Mai (* equal contribution)
EuroSys '27 Spring 2027 Accepted
SwarmPilot introduces agentic scheduling for low-latency agentic applications. It combines scheduling-specific predictors over prompt, device, runtime, and target-model features with a scheduler–agent framework, reducing tail latency by up to 61.5% and sustaining up to 2× the throughput of baseline schedulers under the same SLO in the reported evaluation.
SwarmPilot: Agentic Scheduling for Agentic Systems
Yeqi Huang*, Yanwei Ye*, Guomin Chen, Wenhao Su, Bin Gong, Jialian Li, Zhan Lu, Yangshen Deng, Xuan Sun, Le Xu, Luo Mai (* equal contribution)
EuroSys '27 Spring 2027 Accepted
SwarmPilot introduces agentic scheduling for low-latency agentic applications. It combines scheduling-specific predictors over prompt, device, runtime, and target-model features with a scheduler–agent framework, reducing tail latency by up to 61.5% and sustaining up to 2× the throughput of baseline schedulers under the same SLO in the reported evaluation.
Wavel: A Fast and Efficient Compilation System for Wafer-Scale Accelerators
Yeqi Huang, Congjie He*, Haocheng Xiao*, Yanwei Ye*, Yi-Chieh Wang*, Boyao Song*, Yangshen Deng*, Ziming Miao*, Lingxiao Ma*, Fan Yang*, Luo Mai* (* equal contribution)
SOSP 2026 2026 Accepted
Wavel presents a compilation system for wafer-scale accelerators that makes executable schedules explicit, uses locality-preserving CUT and TUNE primitives, and evaluates physical cost during search. The reported results show higher throughput than expert-optimized WaferLLM and substantial speedups over a Full-Chip baseline on WSE-3 workloads.
Wavel: A Fast and Efficient Compilation System for Wafer-Scale Accelerators
Yeqi Huang, Congjie He*, Haocheng Xiao*, Yanwei Ye*, Yi-Chieh Wang*, Boyao Song*, Yangshen Deng*, Ziming Miao*, Lingxiao Ma*, Fan Yang*, Luo Mai* (* equal contribution)
SOSP 2026 2026 Accepted
Wavel presents a compilation system for wafer-scale accelerators that makes executable schedules explicit, uses locality-preserving CUT and TUNE primitives, and evaluates physical cost during search. The reported results show higher throughput than expert-optimized WaferLLM and substantial speedups over a Full-Chip baseline on WSE-3 workloads.