Assistant Professor · Polytechnique Montréal · DORSAL lab

I study how software behaves when it runs, and how teams keep it healthy as it grows.

My group works on performance and tracing for large heterogeneous systems, on what LLMs can and cannot do for software engineers, and on the everyday machinery of software teams: code review, continuous integration, and APIs.

Publications Team Google Scholar

Portrait of Maxime Lamothe

About

I am an assistant professor in the Department of Computer Engineering and Software Engineering at Polytechnique Montréal, and one of the professors of the DORSAL lab, where I lead the lab's work on heterogeneous and high-performance computing.

Most of what I do is empirical: we mine repositories, traces, logs and review histories from real projects, build tools on top of what we learn, and evaluate them with the people who would use them. Recent work covers GPU benchmarking and tracing, root-cause analysis in microservices, LLM agents for refactoring, and how machine-learning frameworks evolve differently from other software.

Before joining Polytechnique, I was a postdoctoral researcher in the Software REBELs lab at the University of Waterloo with Shane McIntosh, studying build systems and code review. My Ph.D. (Concordia University, 2020, with Weiyi Shang) was about closing the knowledge gap between the developers of APIs and the people who use them. I also hold an M.Eng. from Concordia (2017) and a B.Eng. from McGill (2013).

Active project

DORSAL lab

I am one of the professors of DORSAL, Polytechnique Montréal's lab on the performance and reliability of distributed, cloud and multi-core systems. I supervise its track on heterogeneous and high-performance computing: tracing, profiling and debugging GPU programs at scale.

My work in DORSAL

Recent papers

    All publications