Jiayuan Zhou
jiayuan.zjy[at]gmail.com
Toronto, Canada
Principal Researcher on AI agents for software engineering and vulnerability management.
I am a Principal Researcher at Huawei Canada, leading a team across two complementary lines:
- AI agents for software engineering. Our Lingxi agent framework — #1 on SWE-bench Verified — mines knowledge from development data and agent trajectories to guide the agent harness and evolve the underlying models. Deployed across internal product lines at Huawei.
- OSS vulnerability management. We pushed vulnerability defense from reactive CVE response to proactive sensing — detecting silent fix commits 1–2 weeks ahead of public disclosure through large-scale code-change modeling.
A common thread runs through both: turning implicit, hard-to-observe signals in software development events into explicit systems that can act on them.
I received my Ph.D. in Computer Science from the SAIL lab, Queen’s University, under the supervision of Prof. Ahmed E. Hassan and Prof. Shaowei Wang. My dissertation studied extrinsic incentives in open source communities through mining GitHub, Stack Overflow, and Bountysource data. Before grad school, I was the founding engineer of 1688’s One-Click Dropshipping (一件代发/一键代销) system at Alibaba Group — a full-stack platform connecting B2B suppliers with millions of Taobao/Tmall merchants.
See Research Highlights for a deeper look at these research lines.
Research interests: AI agents for software engineering, procedural knowledge mining, agent memory, development-process knowledge, LLM-based code generation, vulnerability detection and management, mining software repositories.
Publications: 22 papers at ICSE, FSE, ASE, ISSTA, IEEE TSE, ACM TOSEM, and EMSE — see google scholar. I hold 12 patents in software engineering and AI applications.
news
| Apr 16, 2026 | Our paper was accepted by ISSTA 2026: “Answer is Cheap, Show Me the Evidence! Augmenting Automated Vulnerability Assessment with Evidence” |
|---|---|
| Apr 16, 2026 | Our paper was accepted by ISSTA 2026: “Lingxi: Repository-Level Issue Resolution Framework Enhanced by Procedural Knowledge Guided Scaling” |
| Mar 20, 2026 | Our paper was accepted by FSE 2026 (Industry Track): “Mitigating Implicit Inconsistencies in Patch Porting” |
| Mar 18, 2026 | Our paper was accepted by TOSEM: “Detecting Protracted Vulnerabilities in Open Source Projects” |
| Oct 16, 2025 | Our paper was accepted by ICSE’26: “Diffploit: Facilitating Cross-Version Exploit Migration for Open Source Library Vulnerabilities”. See you in Brazil🇧🇷~ |