Xinyu Zeng 曾新宇

I'm the co-founder and CTO of a stealth startup.

I received my Ph.D. in Computer Science from IIIS, Tsinghua University, advised by Huanchen Zhang. I also worked closely with Andy Pavlo. Before that, I graduated with distinction from the University of Wisconsin–Madison with a B.S. in Computer Science and Mathematics, advised by Xiangyao Yu.

My research interests include multi-modal data lakes, data infrastructure for AI, and columnar file formats.

Portrait of Xinyu Zeng

Publications

Google Scholar

  1. WaRP: Workload-Aware Request Planning for Vector Retrieval on Object Storage

    Ruipeng Wan, Xinyu Zeng, Yiming Qiao, Huanchen Zhang

    VLDB 2027 Proceedings of the VLDB Endowment

  2. G-Casdec: General Cascaded Decompression on GPUs

    Yongqi Zhuo*, Xinyu Zeng*, Huanchen Zhang, Mingyu Gao

    SIGMOD 2027 Proceedings of the ACM on Management of Data

    [Paper][Code]

  3. F3: The Open-Source Data File Format for the Future

    Xinyu Zeng, Ruijun Meng, Martin Prammer, Wes McKinney, Jignesh M. Patel, Andrew Pavlo, Huanchen Zhang

    SIGMOD 2026 Proceedings of the ACM on Management of Data

    Best Paper Honorable Mention

    [Paper][Code][Slides][Podcast]

  4. Towards Functional Decomposition of Storage Formats

    Martin Prammer, Xinyu Zeng, Ruijun Meng, Wes McKinney, Huanchen Zhang, Andrew Pavlo, Jignesh M. Patel

    CIDR 2025 Proceedings of the 2025 Conference on Innovative Data Systems Research

    [Paper]

  5. NULLS! Revisiting Null Representation in Modern Columnar Formats

    Xinyu Zeng, Ruijun Meng, Andrew Pavlo, Wes McKinney, and Huanchen Zhang

    DaMoN 2024 Proceedings of the 20th International Workshop on Data Management on New Hardware

    [Paper][Code][Slides]

  6. LeCo: Lightweight Compression via Learning Serial Correlations

    Yihao Liu, Xinyu Zeng, Huanchen Zhang

    SIGMOD 2024 Proceedings of the ACM on Management of Data

    [Paper][arXiv][Code]

  7. An Empirical Evaluation of Columnar Storage Formats

    Xinyu Zeng, Yulong Hui, Jiahong Shen, Andrew Pavlo, Wes McKinney, Huanchen Zhang

    VLDB 2023 Proceedings of the VLDB Endowment

    [Paper][Extended version][Code][Slides]

  8. 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 Proceedings of the VLDB Endowment

    [Paper][arXiv][Code]

Experience

Education

Awards

Teaching

Tsinghua University

  • Teaching Assistant — Database Systems

    Spring 2023

University of Wisconsin–Madison