Biography I am Qiaolin Yu, a graduate student at Cornell University and Cornell Tech. My interests primarily focus on LSM-Tree based key-value stores, database systems, machine learning systems, and haircutting. Currently, I work as a research intern at Cornell Computer Systems Laboratory, advised by Prof. Udit Gupta. I also work closely with Prof. Zhichao Cao. I also have amassed significant industry experience. I was a software engineer intern at Databricks in 2024 summer. Before that, I interned in the compute architecture team at PingCAP, the Azure Machine Learning team at Microsoft, the ByteHouse team at ByteDance, and the graph database team at Kuaishou. I will graduate in May 2025 and am actively seeking NG opportunities related to quant, LLM, and systems. If you would like to chat, feel free to reach out. My resume is available upon request. Mail     GitHub      LinkedIn      Google Scholar

News

[2024/02] My first-author paper was accepted by SIGMOD 2024. Super excited!!!

[2023/09] Got the 2024 summer internship offer from Databricks. See you all in Mountain View!

Education

Cornell University (Cornell Tech) Aug 2023 - May 2025

MS, Computer Science and Information Systems

University of Liverpool (Xi'an Jiaotong-Liverpool University) Sep 2019 - Jul 2023

Industry Experience

Databricks May 2024 - Aug 2024

• Built a simulator for the auto-sharder.

PingCAP Aug 2022 - May 2023

• Developed the auto-scaling feature for TiFlash (the analytical engine for TiDB) from the ground up.

Microsoft May 2022 - Aug 2022

• Built infrastructure for Azure Machine Learning service.

ByteDance Jan 2022 - Apr 2022

• Worked on ByteHouse (also known as ByConity within the open-source community), a cloud native data warehouse.

Kuaishou Jun 2021 - Sep 2021

• Designed and developed infrastructure for NebulaGraph clusters.

Research Experience

Computer Systems Laboratory, Cornell University Jan 2024 - Present

• Advised by Prof. Udit Gupta.
• Optimizing data I/O performance in end-to-end recommendation systems.

Intelligent Data Infrastructure Lab, Arizona State University Dec 2021 - Present

• Advised by Prof. Zhichao Cao.
• Conducting research on RocksDB and its related database systems, such as MyRocks, NebulaGraph, and Kvrocks, with a focus on their performance.

Publications

CaaS-LSM: Compaction-as-a-Service for LSM-based Key-Value Stores in Storage Disaggregated Infrastructure Qiaolin Yu, Chang Guo, Jay Zhuang, Viraj Thakkar, Jianguo Wang, Zhichao Cao. ACM Conference on Management of Data (SIGMOD 2024), Research Track Full Paper. [paper] [code]

	Prophet: Optimizing LSM-Based Key-Value Store on ZNS SSDs with File Lifetime Prediction and Compaction Compensation Gaoji Liu, Chongzhuo Yang, Qiaolin Yu, Chang Guo, Wen Xia, Zhichao Cao. International Conference on Massive Storage Systems and Technology (MSST 2024), Research Track Full Paper. [paper] [code]

Invited Talks

• "Optimizing LSM-based Key-Value Stores in Storage Disaggregated Infrastructure Through Compaction-as-a-Service", Salesforce Database Group, Invited Talk, California, USA, 2025
• "CaaS-LSM: Compaction-as-a-Service for LSM-based Key-Value Stores in Storage Disaggregated Infrastructure", ACM Conference on Management of Data (SIGMOD 2024), Santiago, Chile, 2024
• "Prophet: Optimizing LSM-Based Key-Value Store on ZNS SSDs with File Lifetime Prediction and Compaction Compensation", International Conference on Massive Storage Systems and Technology (MSST 2024), California, USA, 2024

Academic Service

• Reviewer: NAACL 2025, ACL Rolling Review 2024, ICLR 2025 Workshop LLM Reason and Plan, ICLR 2025 Workshop Foundation Models in the Wild
• Availability & Reproducibility Committee: ACM SIGMOD 2024

Teaching

• Grader of CS 5781 Machine Learning Engineering Fall 2024

Miscellaneous