I’m a Member of Technical Staff at xAI. I work on Language Model serving. I’m interested in Machine Learning Systems, Caching, and Distributed Systems.

I was a researcher at SystemsResearch@Google. I worked on Language Model Serving and Memory Management. I worked with David Culler and Hank Levy, and collaborating with teams across Cloud and DeepMind. I contributed to Gemma 3n, gemma.cpp, and LLVM.

I completed my Ph.D. at the University of Wisconsin-Madison in 2022, advised by Andrea and Remzi Arpaci-Dusseau. I obtained my bachelor’s degree from the University of Science and Technology of China (USTC) in 2016.

Experience

Member of Technical Staff
xAI
2025 - Now

Senior Software Engineer
SystemsResearch@Google
2022 - 2025

Affiliate Research Assistant
Microsoft Gray System Lab
2018 - 2021

Software Engineering Intern
VMware
2019 Summer

Undergrad Research Assistant
Chinese University of Hong Kong
2016

Publications

Spark Transformer: Reactivating Sparsity in FFN and Attention
Chong You*, Kan Wu*(co-first author), Zhipeng Jia*, Lin Chen*, Srinadh Bhojanapalli, Jiaxian Guo, Utku Evci, Jan Wassenberg, Praneeth Netrapalli, Jeremiah J. Willcock, Suvinay Subramanian, Felix Chern, Alek Andreev, Shreya Pathak, Felix Yu, Prateek Jain, David E. Culler, Henry M. Levy, Sanjiv Kumar
ArXiv 2025: [Preprint]

Getting the MOST out of your Storage Hierarchy with Mirror-Optimized Storage Tiering
Kaiwei Tu, Kan Wu, Andrea Arpaci-Dusseau, Remzi Arpaci-Dusseau
FAST’2026: 24th USENIX Conference on File and Storage Technologies [to appear]

PageFlex: Flexible and Efficient User-space Delegation of Linux Paging Policies with eBPF
Anil Yelam, Kan Wu*(corresponding author), Zhiyuan Guo, Rajath Shashidhara, Stanko Novakovic, Suli Yang, Wei Xu, Alex C. Snoeren, Kimberly Keeton
ATC’2025: 2025 USENIX Annual Technical Conference [to appear]

FineMem: Breaking the Allocation Overhead vs. Memory Waste Dilemma in Fine-Grained Disaggregated Memory Management
Xiaoyang Wang, Yongkun Li, Kan Wu, Wenzhe Zhu, Yuqi Li, Yinlong Xu
OSDI’2025: 19th USENIX Symposium on Operating Systems Design and Implementation [to appear]

SLAP: Segmented Reuse-Time-Label Based Admission Policy for Content Delivery Network Caching
Ke Liu, Kan Wu, Hua Wang, Ke Zhou, Peng Wang, Ji Zhang, Cong Li
TACO’2024: ACM Transactions on Architecture and Code Optimization [paper]

Arya: Arbitrary Graph Pattern Mining with Decomposition-based Sampling
Kan Wu*(co-first author), Zeying Zhu*, Zaoxing Liu
NSDI’2023: 20th USENIX Symposium on Networked Systems Design and Implementation [paper]

SLAP: An Adaptive, Learned Admission Policy for Content Delivery Network Caching
Ke Liu, Kan Wu, Hua Wang, Ke Zhou, Ji Zhang, Cong Li
IPDPS’2023: 37th International Parallel and Distributed Processing Symposium [paper]

WiscSort: External Sorting for Byte Addressable Storage
Vinay Banakar, Kan Wu, Yuvraj Patel, Kimberly Keeton, Andrea Arpaci-Dusseau, Remzi Arpaci-Dusseau
VLDB’2023: 49th International Conference on Very Large Data Bases [paper]

NyxCache: Flexible and Efficient Multi-tenant Persistent-Memory Caching
Kan Wu, Kaiwei Tu, Yuvraj Patel, Rathijit Sen, Kwanghyun Park, Andrea Arpaci-Dusseau, Remzi Arpaci-Dusseau
FAST’2022: 20th USENIX Conference on File and Storage Technologies [paper] [slides] [video]

Cornus: Atomic Commit for Cloud DBMS with Storage Disaggregation
Zhihan Guo, Xinyu Zeng, Kan Wu, Wuh-Chwen Hwang, Ziwei Ren, Xiangyao Yu, Mahesh Balakrishnan, Philip A. Bernstein
VLDB’2022: 48th International Conference on Very Large Data Bases [paper]

The Storage Hierarchy is Not a Hierarchy: Optimizing Caching on Modern Storage Devices with Orthus
Kan Wu, Zhihan Guo, Guanzhou Hu, Kaiwei Tu, Ramnatthan Alagappan, Rathijit Sen, Kwanghyun Park, Andrea Arpaci-Dusseau, Remzi Arpaci-Dusseau
FAST’2021: 19th USENIX Conference on File and Storage Technologies [paper] [slides] [video] [code]

The Storage Hierarchy is Not a Hierarchy: Optimizing Caching on Modern Storage Devices with Orthus
Kan Wu et al.
NVMW’2021: 12th Non-Volatile Memories Workshop [paper]

Releasing Locks As Early As You Can: Reducing Contention of Hotspots by Violating Two-Phase Locking
Zhihan Guo, Kan Wu, Cong Yan, Xiangyao Yu
SIGMOD’2021: ACM SIGMOD International Conference on Management of Data [paper]

Read as Needed: Building WiSER, a Flash-Optimized Search Engine
Jun He, Kan Wu, Sudarsun Kannan, Andrea Arpaci-Dusseau, Remzi Arpaci-Dusseau
FAST’2020: 18th USENIX Conference on File and Storage Technologies [paper]

Towards an Unwritten Contract of Intel Optane SSD
Kan Wu, Andrea Arpaci-Dusseau, Remzi Arpaci-Dusseau
HotStorage’2019: 11th USENIX Workshop on Hot Topics in Storage and File Systems [paper] [slides] [code]

Exploiting Intel Optane SSD for Microsoft SQL Server
Kan Wu, Andrea Arpaci-Dusseau, Remzi Arpaci-Dusseau, Rathijit Sen, Kwanghyun Park
DaMoN, SIGMOD’2019: ACM SIGMOD International Conference on Management of Data [paper]

Spark Transformer: Reactivating Sparsity in FFN and Attention
Chong You, Kan Wu, Zhipeng Jia, Lin Chen, Srinadh Bhojanapalli, Jiaxian Guo, Utku Evci, Jan Wassenberg, Praneeth Netrapalli, Jeremiah J. Willcock, Suvinay Subramanian, Felix Chern, Alek Andreev, Shreya Pathak, Felix Yu, Prateek Jain, David E. Culler, Henry M. Levy, Sanjiv Kumar
arXiv’2025: arXiv:2506.06644 [paper]

Service

DIMES’25 PC
USENIX FAST’25 PC
USENIX ATC’25 PC
USENIX ATC’24 PC
USENIX HotStorage’24 PC
SOSP’24 Proceedings Chair
SOSP’23 Student Research Committee, Poster Committee
ACM TOS’2021, 2022, 2023, 2024 reviewer
VLDBJ’2023 reviewer