Abdulrahman Diaa

Hello, my name is Abdulrahman Diaa. I am a Ph.D. candidate under the supervision of Florian Kerschbaum. I'm also a member of the Cryptography, Security, and Privacy (CrySP) lab at the University of Waterloo. I hold a master's degree in Computer Science from the University of Waterloo and a Bachelor of Science in Computer Engineering and Mathematics from The American University in Cairo.

My research addresses privacy and traceability in machine learning. I design efficient privacy-preserving protocols that enable collaborative analytics without exposing sensitive data, using secure multi-party computation and differential privacy. I also study robust watermarking for AI-generated content, exposing vulnerabilities in existing schemes through adaptive attacks and building defenses that withstand adversarial removal.

Currently, I am a student researcher at Google working on private RAG and nearest neighbor search with Sarvar Patel. Previously, I worked with Erik-Oliver Blass at Airbus on privacy-preserving DBSCAN clustering, and with Jean-Pierre Hubaux at EPFL on multi-party homomorphic encryption for federated learning.

In my free time, I take coffee a bit too seriously. My current V60 Switch recipe has its own page, naturally. My favorite coffee (and life) companion will always be Monireh.