My research interests broadly cover security and privacy aspects of large online services. I am particularly interested in the emerging space at the intersection of machine learning and security. Three research directions include: (1) Attacks using ML: Investigating how recent developments in ML can be misused to abuse online services. (2) Attacks on ML systems: Understanding vulnerabilities of ML systems and improving robustness. (3) ML for better security/privacy: Leveraging advances in ML to build better defenses against attacks on online services.