About me

I am a researcher at the Vector Institute, working on privacy (techniques, concerns, policies) in machine learning and data management. My current projects are on privacy-preserving federated learning and synthetic data generation for finance and health data.

I have worked on a variety of topics in security, privacy, and cryptography over the past decade in academia and industry. My main interests are: security and privacy in machine learning/data management, differential privacy, mutli-party computation, and applied cryptography.

I received my PhD from the University of Waterloo in 2021, on data protection in big data analysis. During my PhD and in Fall 2020, I worked as a research intern at National Research Council (NRC) in Canada. My research project was designing a secure mechanism to perform join operation over encrypted data. I conducted another internship over Summer 2020, at Microsoft Research (MSR) in the US, focusing on privacy techniques (in particular differential privacy) for correlated data, while training models on a graph of organizational communications (e.g. emails).

After my PhD, I worked as a senior cryptography consultant at Royal Bank of Canada. At RBC, I lead preparation for post-quantum cryptography migration, consulted with technology groups, and developed cryptographic architectural patterns and standards. Noticing the lack of a successful deployment of privacy protecting techniques in industry, I took a leave from RBC and joined the Vector Institute to research on the impedences on this deployment and the ways to facilitate it for various industry sectors. I welcome any discussion and collaboration on the topic.