Amine Barrak

PhD at University
of Québec


I am Amine Barrak, an Assistant Professor at Oakland University. I hold a Ph.D. in Computer Science from the from the University of Quebec, specifically from the Department of Computer Science and Mathematics. My research focuses on adapting distributed training in machine learning to function-as-a-service (FaaS) infrastructure.

Research Interests

  • Distributed ML Architectures
  • Function-as-a-Service (FaaS) and Cloud Computing
  • Software Engineering for ML (SE4ML)
  • MLOps and ML Pipeline
  • Secure Machine Learning
  • Parallel Algorithms

For details about my research, please check my publication list.

I obtained my research master’s at Polytechnique Montréal, where I applied machine learning techniques to detect vulnerabilities in source code. My work was inspired by concerns over unintentional privilege protection changes that could introduce security vulnerabilities and expose sensitive data. My master’s thesis won the Best Student Paper Award. This work was supervised by Foutse Khomh. As a research assistant with Bram Adams, I contributed to continuous integration and machine learning pipelines, analyzing source code and build failures while working on the traceability of ML and source code artifacts.

I completed my engineering degree in Tunisia at the Higher Institute of Informatics, where I earned a government scholarship based on my exceptional academic achievements, enabling me to pursue studies in Canada.

Latest News

  • Aout 2024: I got the position of assiatnt professor at Oakland University.
  • Jun 2024: Accepted paper intitled: “Securing AWS Lambda: Advanced Strategies and Best Practices”, CSCLOUD 2024 (Acceptance Rate: 24.66%)
  • May 2024: The poster for the PhD Forum at IPDPS 2024 is ready to be shared link
  • May 2024: Teaching the Cryptography course for two groups of students totaling 54 students. Course details.
  • May 2024: Organizing the Journée Cyberdéfense Desjardins at the University of Quebec with my research mentor Fehmi Jaafar.


Selected Publications

  • Serverless on machine learning: A systematic mapping study
    IEEE Access 2022
  • Why do builds fail?—A conceptual replication study
    Journal of Systems and Software 2021
  • SPIRT: A Fault-Tolerant and Reliable Peer-to-Peer Serverless ML Training Architecture
    2023 IEEE 23rd International Conference on Software Quality, Reliability and Security (QRS)
  • On the Co-evolution of ML pipelines and source code-empirical study of DVC projects
    2021 IEEE International Conference on Software Analysis, Evolution and Reengineering (SANER)