Qiyang He

Member of Technical Staff
Microsoft AI — Agentic Coding RL Team
Contact: qiyanghe1998 AT outlook DOT com
Research Interests
  • Agentic Coding & Reinforcement Learning
  • AI Training & Post-Training Infrastructure
  • Database Systems & Query Optimization
  • Scalable AI & Distributed Data Systems

I am Qiyang He (贺启旸), a Member of Technical Staff at Microsoft AI on the Agentic Coding RL Team. I build scalable infrastructure for coding reinforcement learning, turning real-world software engineering tasks and agent trajectories into reproducible, executable environments for training and evaluation.

I received my M.Sc. in Computer Science from Purdue University, where I was advised by Prof. Tiark Rompf. Before that, I earned a B.Eng. from the Department of Computer Science and Engineering at Southern University of Science and Technology (SUSTech), advised by Prof. Bo Tang.

My interests span agentic coding systems, reinforcement learning and post-training infrastructure, database systems, and distributed systems. Previously, I was a Software Engineer on the Metadata Team at Snowflake, working on scalable metadata management for petabyte-scale analytical queries. I also interned at Cockroach Labs and Pinterest.

Education
GPA: 3.91/4.00
West Lafayette, IN, USA
GPA: 3.86/4.00, Rank: 3/146
Advisor: Prof. Bo Tang
Shenzhen, Guangdong, China
Experience
Member of Technical Staff | Agentic Coding RL Team
Software Engineer | Metadata Team
Research Intern | SQL Query Team
Research Intern | Distributed Storage System Team
Research Assistant | Programming Language Group
Research Assistant | Database Group
Advisor: Prof. Bo Tang
Selected Publications
MAI-Thinking-1: Building a Hill-Climbing Machine

The Microsoft AI Team (SWE RL environments)
[pdf]

A Survey of Learned Indexes for the Multi-dimensional Space

Abdullah Al-Mamun, Hao Wu, Qiyang He, Jianguo Wang, Walid G. Aref
ACM Computing Surveys (ACM CSUR), 2025
[pdf] [acm dl]

Efficient Incrementalization of Correlated Nested Aggregate Queries using Relative Partial Aggregate Indexes (RPAI)

Supun Abeysinghe, Qiyang He, Tiark Rompf
Proceedings of the 2022 International Conference on Management of Data (SIGMOD 2022). Philadelphia, PA, USA
[pdf] [acm dl]

Reachability Types: Tracking Aliasing and Separation in Higher-Order Functional Programs

Yuyan Bao, Guannan Wei, Oliver Bračevac, Yuxuan Jiang, Qiyang He, Tiark Rompf
Proceedings of the ACM on Programming Languages, Volume 5 (OOPSLA 2021). Online/Chicago, IL, USA
[pdf] [acm dl] [artifact]