Boqi (Percy) Chen, Ph.D.

McGill University

boqi_chen.jpg

email: boqi⚫chen@mail⚫mcgill⚫ca

✨ I will be joining the University of Ottawa as a tenure-track Assistant Professor in the School of Electrical Engineering and Computer Science (EECS) starting May 2026! If you are interested in working with me, please feel free to reach out!

I received my PhD from the Department of Electrical and Computer Engineering at McGill University, supervised by Prof. Gunter Mussbacher and Prof. Daniel Varro. My research focuses on the reliable and robust integration of AI components, including LLMs, in software engineering applications, with an emphasis on model-based validation techniques and principled agentic workflow design. I am also broadly interested in evaluating the quality and properties of ML models for software engineering tasks such as code generation, bug detection, and code summarization.

During my PhD, I worked as an R&D engineer at Aggregate Intellect through a Mitacs grant and as a part-time research associate at Huawei Waterloo Research Center. I was also an active contributor to the open-source project Sherpa, a large language model framework for robust agentic applications.

I believe reliable and robust integration of AI components requires a layered approach, as illustrated below.

System Layer

How to integrate AI components into software systems in a reliable and robust way, including reliable workflow construction (MODELS'25) and system-level evaluation (MODELS'25, ICSE'25).

Traditional Software Component

Create reliable software components that wrap around AI components to provide assurance, guarantees and fallback mechanisms (ASE'22, RE'25, MODELS'25)

AI Component

How to provide reliability and robustness guarantees by design for AI components, including prompting, agentic architecture, and decoding strategies.

Model Layer

Open the black box of AI models to understand their properties and behaviors (DMKD 2026) or improve the neural architectures (FSE'25).

I also organize a bi-weekly (sometimes weekly) discussion group on graph neural networks and their applications. You can find the scedules here. If you are interested in joining, please subscribe to this calendar. We are also actively looking for new speakers, so if you are interested in presenting your work or any topics you are interested in, please feel free to reach out to me.

news

Dec 12, 2025 My work “Certifying robustness of graph convolutional networks for node perturbation with polyhedra abstract interpretation” has been accepted to Data Mining and Knowledge Discovery (DMKD) journal after a long revision process! Check the paper out here.
Oct 20, 2025 I’ve successfully defended my PhD thesis titled “Domain-driven and Consistent Integration of Large Language Models: An Input-Output Perspective”!
Jul 04, 2025 Three papers that I’ve contributed to have been accepted to the ACM / IEEE 28th International Conference on Model Driven Engineering Languages and Systems (MODELS 2025).

selected publications

  1. MODELS
    chen2025sherpa.png
    SHERPA: A Model-Driven Framework for Large Language Model Execution
    Boqi Chen, Kua Chen, José Antonio Hernández López , and 3 more authors
    In ACM / IEEE 28th International Conference on Model Driven Engineering Languages and Systems (MODELS), 2025, Grand Rapids, USA, October 5-10, 2025 , 2025
  2. MODELS
    ahmed2025mcet.png
    MCeT: Behavioral Model Correctness Evaluation using Large Language Models
    Khaled Ahmed, Jialing Song, Boqi Chen , and 2 more authors
    In ACM / IEEE 28th International Conference on Model Driven Engineering Languages and Systems (MODELS), 2025, Grand Rapids, USA, October 5-10, 2025 , 2025
  3. MODELS
    chen2025accurate.png
    Accurate and Consistent Graph Model Generation from Text with Large Language Models
    Boqi Chen, Ou Wei, Bingzhou Zheng , and 1 more author
    In ACM / IEEE 28th International Conference on Model Driven Engineering Languages and Systems (MODELS), 2025, Grand Rapids, USA, October 5-10, 2025 , 2025
  4. ICSE
    chen2025power.jpg
    The Power of Types: Exploring the Impact of Type Checking on Neural Bug Detection in Dynamically Typed Languages
    Boqi Chen, José Antonio Hernández López, Gunter Mussbacher , and 1 more author
    In Proceedings of the 2025 IEEE/ACM 47th International Conference on Software Engineering , 2025
  5. MODELS
    chen2023automated.jpg
    Automated domain modeling with large language models: A comparative study
    Kua Chen, Yujing Yang, Boqi Chen , and 3 more authors
    In 2023 ACM/IEEE 26th International Conference on Model Driven Engineering Languages and Systems (MODELS) , 2023