Moshi Wei

AI / Software Engineering researcher and engineer working at the intersection of AI for Software Engineering, applied large language model systems, agent infrastructure, developer tools, and software reliability. Research experience includes API recommendation for machine-learning libraries, deep-learning API misuse detection, neural code-model testing, code generation, and static analysis. Professional work includes production LLM systems, workflow automation, agent orchestration, persistent assistant runtimes, and open-source developer tooling.

Senior Machine Learning Engineer at LRQA (A Goldman Sachs Company) (2024-06 – Present)

• Led production LLM projects across scheduling optimization, audit-report generation, and automated audit-report review.
• Built LLM-based tools for audit-report generation, automated audit review, and scheduling optimization.

Software Engineering Researcher Intern at Huawei Research Canada (2023-09 – 2024-10)

• Conducted software-engineering research on LLM- and agent-based systems, including a Paper Reading Agent for literature and lead-discovery workflows.

Machine Learning Intern at Huawei Canada Research Institute (2023-01 – 2023-04)

• Worked on machine-learning methods for mobile-edge computing and split learning.
• Co-authored a GLOBECOM 2023 paper on dynamic encoding and decoding for split learning in mobile-edge computing with Omar Alhussein and collaborators.

Software Engineer at Achievers (2019-08 – 2021-07)

• Led an award-winning machine-learning recommender-system project.
• Built natural-language processing components for named-entity recognition.
• Reworked a PostgreSQL-based data-warehouse restoration workflow.

Ph.D. in Software Engineering – York University (2020-09 – 2025-05)

Master of Applied Science (MASc) in Electrical and Computer Engineering – University of Waterloo (2017-12 – 2019-05)

B.Sc. in Computer Science – University of Ottawa (2012-09 – 2016-05)