About Me

Hi, my name is Yin Lin. I am a senior algorithm engineer at Tongyi Lab, Alibaba Group. My current research focuses on agent harness design, particularly memory systems and reasoning paradigms that enable agents to manage context, reason effectively, and execute long-horizon tasks.

Our team develops QwenPaw QwenPaw GitHub stars, a secure and extensible personal AI assistant that runs locally or in the cloud and uses Scroll Context to persist and index its full interaction history for on-demand recall. Alongside my work on agent harnesses, I also develop techniques for AI-powered data management, including semantic operators and AI agents that enhance data processing, analytics, and reasoning.

Before joining Alibaba, I earned my Ph.D. from the University of Michigan, Ann Arbor, where I was advised by Dr. H. V. Jagadish. My thesis research focused on data equity systems.

If you are interested in a research internship or collaboration, feel free to reach out!

What's New
  • QwenPaw Version 2.0 brings major upgrades to its agent architecture and capabilities.
  • VLDB 2026 Our tutorial on LLM-enhanced relational operators is accepted at VLDB 2026.
  • SIGMOD 2026 Our demo on ambiguity resolution for text-to-SQL is accepted at SIGMOD 2026.
  • Data-Juicer Semantic operators are now available for large-scale training data processing.
  • DojoZero An agent arena for real-time data stream prediction in sports events. View AI agents compete in realtime at dojozero.live.

Education & Experience

Education

Sept. 2019 – Dec. 2024

University of Michigan

Ph.D., Computer Science and Engineering

Sept. 2015 – June 2019

Shanghai Jiao Tong University

B.S., Computer Science

Experience

Dec. 2024 – Present

Alibaba Group

Senior Algorithm Engineer, Tongyi Lab

June 2022 – Aug. 2022

Microsoft Research

Research Intern, Data Management, Exploration and Mining (DMX)

May 2018 – Jul. 2018

University of Waterloo

Summer Intern, Software Architecture Group

Research Interest

I am currently exploring how AI agents can better manage, reason over, and interact with data. Here are some of my ongoing projects:

Ambiguity Resolution in Analytical Queries

SIGMOD 2026 Demo

Natural language queries to databases are often ambiguous. AmbiSQL is an interactive system that detects and resolves ambiguities in text-to-SQL translation, enabling users to clarify their intent and receive accurate query results.

Semantic Operators for Data Processing

VLDB 2026 Tutorial

Semantic operators bring AI-powered data transformations to large-scale data processing pipelines. These operators are now part of Data-Juicer, enabling intelligent filtering, enrichment, and transformation of training data for foundation models.

DojoZero: Agent Arena

GitHub Open-Source Project

DojoZero is an agent arena where AI agents react to real-time data streams to make predictions for sports events. It serves as a testbed for evaluating agents' capabilities in dynamic, time-sensitive decision-making scenarios.

AI-Driven Data Analytics

IEEE Data Engineering Bulletin 2025 · CIDR 2024

Exploring novel agent applications for data analytics, including using LLMs for feature engineering and data engineering tasks.

Publications

1

Bridging LLMs and Database Systems: A Deep Dive into Enhanced Relational Operators

Tianjing Zeng, Yin Lin, Rong Zhu, Yunxiang Su, Zhongjun Ding, Bolin Ding, Jingren Zhou

VLDB 2026 (Tutorial)

2

AmbiSQL: Interactive Ambiguity Detection and Resolution for Text-to-SQL [pdf]

Zhongjun Ding, Yin Lin*, Tianjing Zeng, Rong Zhu, Bolin Ding, Jingren Zhou (* corresponding author)

SIGMOD 2026 (Demo)

3

Large Language Models as Pretrained Data Engineers: Techniques and Opportunities [pdf]

Yin Lin, Bolin Ding, Jingren Zhou

IEEE Data Engineering Bulletin 2025

4

Efficient Row-Level Lineage Leveraging Predicate Pushdown [pdf]

Yin Lin, Cong Yan

CoRR, 2024, Arxiv/2412.16864

5

Mitigating Subgroup Unfairness in Machine Learning Classifiers: A Data-Driven Approach [pdf]

Yin Lin, Samika Gupta, H. V. Jagadish

ICDE 2024

6

SMARTFEAT: Efficient Feature Construction through Feature-Level Foundation Model Interactions [pdf]

Yin Lin, Bolin Ding, H. V. Jagadish, Jingren Zhou

CIDR 2024

7

Predicate Pushdown for Data Science Pipelines [pdf]

Cong Yan, Yin Lin, Yeye He

SIGMOD 2023 (Best Paper Award)

8

Representation Bias in Data: A Survey on Identification and Resolution Techniques [pdf]

Nima Shahbazi, Yin Lin, Abolfazl Asudeh, H. V. Jagadish

ACM Computing Surveys

Highly cited survey in the field of AI fairness
9

OREO: Detection of Cherry-picked Generalizations [pdf]

Yin Lin, Brit Youngmann, Yuval Moskovitch, H. V. Jagadish, Tova Milo

VLDB 2022 (Demo)

10

On Detecting Cherry-picked Generalizations [pdf]

Yin Lin, Brit Youngmann, Yuval Moskovitch, H. V. Jagadish, Tova Milo

VLDB 2022

11

Identifying Insufficient Data Coverage in Databases with Multiple Relations [pdf]

Yin Lin, Yifan Guan, Abolfazl Asudeh, H. V. Jagadish

VLDB 2020

12

MithraDetective: A System for Cherry-picked Trendlines Detection [pdf]

Yoko Nagafuchi, Yin Lin, Kaushal Mamgain, Abolfazl Asudeh, H. V. Jagadish, You (Will) Wu, Cong Yu

CoRR, 2020, Arxiv/2010.08807

13

On Structural vs. Proximity-based Temporal Node Embeddings [pdf]

Puja Trivedi, Alican Büyükçakır, Yin Lin, Yinlong Qian, Di Jin, Danai Koutra

MLG@KDD 2020

14

R2-Tree: An Efficient Indexing Scheme for Server-Centric Data Center Networks [pdf]

Yin Lin, Xinyi Chen, Xiaofeng, Guihai Chen

DEXA 2018

Scholarships and Awards

Best Paper Award at SIGMOD 2023
NSF Travel Award for ICDE 2024
Rackham Dean's and Named PhD Fellowship, University of Michigan
Outstanding Undergraduate, Shanghai Jiao Tong University
Chun Tsung Scholar, Shanghai Jiao Tong University

Professional Service

Program Committees / Reviewers: ACL, NeurIPS, TKDE, CIKM, IEEE BigData, AIBSD (AAAI Workshop on AI with Biased or Scarce Data), ReLM (AAAI Workshop on Responsible Language Models)