Haifeng Ruan

photo.jpg

haifeng.ruan.office@gmail.com

Hi! My name is Haifeng RUAN (阮海峰). I am a Postdoctoral Research Fellow in agentic AI at the National University of Singapore (NUS). I received my PhD in computer science from NUS in September 2026, advised by Prof. Abhik Roychoudhury.

My research interest lies in improving software quality and trustworthiness, by means of program testing, analysis, and repair. I am particularly interested in improving AI-generated code and facilitating its integration into production systems. For this purpose, I am actively developing AI coding agents that have a deep understanding of program semantics. In the past, I have also worked on finding security vulnerabilities by fuzzing.

I am always open to collaborations and discussions. Feel free to email me for a chat! :)

news

Jun 25, 2026 Our paper AutoCodeSherpa: Symbolic Explanations in AI Coding Agents has been accepted to ISSTA 2026.
May 21, 2026 Sonar has globally launched the SonarQube Remediation Agent at ATxSummit 2026. I am excited to see our research on AutoCodeRover evolve from an NUS project into a commercial product that automatically fixes and verifies code issues for developers worldwide.
Feb 19, 2026 Sonar Foundation Agent, for which I was one of the two main designers at Sonar, claimed the top spot on the unfiltered SWE-bench leaderboard, scoring 79.2% on SWE-bench Verified and 52.62% on SWE-bench Full. It builds on AutoCodeRover with a more autonomous, test-driven workflow for real-world software issue remediation. [Press Release]
Feb 20, 2025 AutoCodeRover, our startup from NUS, has been acquired by Sonar. [Press Release] [Our Team]
Oct 31, 2024 Our paper SpecRover: Code Intent Extraction via LLMs has been accepted to ICSE’25. SpecRover (aka AutoCodeRover-v2.0) guides an LLM agent with program specifications inferred from multiple sources and is the best-performing open-source agent on SWE-Bench!
Jul 03, 2024 Our paper AutoCodeRover: Autonomous Program Improvement has been accepted to ISSTA’24.
Apr 08, 2024 AutoCodeRover, our autonomous software engineering agent, is released! AutoCodeRover performs program structure-aware code search to efficiently retrieve relevant code context. Check out our arXiv report, source code, website, and tweet for details!

publications

  1. ISSTA26
    AutoCodeSherpa: Symbolic Explanations in AI Coding Agents
    Sungmin Kang, Haifeng Ruan, and Abhik Roychoudhury
    In the ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA), 2026
    To appear in ISSTA 2026
    arXiv
  2. ICSE25
    SpecRover: Code Intent Extraction via LLMs
    Haifeng Ruan*, Yuntong Zhang*, and Abhik Roychoudhury
    * joint first authors, ordered alphabetically
    In the 47th International Conference on Software Engineering (ICSE), 2025
    arXiv PDF Code
  3. TOSEM24
    Timing Side-Channel Mitigation via Automated Program Repair
    Haifeng Ruan, Yannic Noller, Saeid Tizpaz-Niari, Sudipta Chattopadhyay, and Abhik Roychoudhury
    ACM Trans. Softw. Eng. Methodol., 2024
    DOI PDF Code
  4. ISSTA24
    Oracle-Guided Program Selection from Large Language Models
    Zhiyu Fan*, Haifeng Ruan*, Sergey Mechtaev, and Abhik Roychoudhury
    * joint first authors, ordered alphabetically
    In Proceedings of the 33rd ACM SIGSOFT International Symposium on Software Testing and Analysis, ISSTA 2024, Vienna, Austria, September 16-20, 2024, 2024
    DOI PDF Code
  5. ICST24
    Evolutionary Testing for Program Repair
    Haifeng Ruan, Hoang Lam Nguyen, Ridwan Shariffdeen, Yannic Noller, and Abhik Roychoudhury
    In IEEE Conference on Software Testing, Verification and Validation, ICST 2024, Toronto, ON, Canada, May 27-31, 2024, 2024
    DOI PDF Code
  6. ISSTA24
    AutoCodeRover: Autonomous Program Improvement
    Yuntong Zhang, Haifeng Ruan, Zhiyu Fan, and Abhik Roychoudhury
    In Proceedings of the 33rd ACM SIGSOFT International Symposium on Software Testing and Analysis, ISSTA 2024, Vienna, Austria, September 16-20, 2024, 2024
    DOI PDF Code