Research Interests
I am broadly interested in applying mathematics to difficult real-world problems. My work spans scientific computing, optimization, operations research, and machine learning — currently focused on privacy risk quantification for sensitive data, with prior work in drug discovery, energy market modelling, and production scheduling.
Current Work
I am currently developing algorithms for privacy risk assessment of de-identified, anonymized, and synthetic data. The work involves quantifying re-identification and inference risks against recognized privacy standards, with an emphasis on implementations that are efficient enough for production use.
Publications
- Multi-Objective Reinforcement Learning for Generating Covalent Inhibitor Candidates arXiv
2026 · arXiv.org
- Graph Neural Networks for Identifying Protein-Reactive Compounds Journal
2024 · Digital Discovery
Also presented at
Conference2023 · NeurIPS 2023
- cclib 2.0: An updated architecture for interoperable computational chemistry. Journal
2024 · Journal of Chemical Physics
- How AI is Changing Chemical Discovery Article
2022 · The Gradient
Teaching
- EPI 6101 — Course Lectures on Neural Networks
2024 – present · University of Ottawa