Senior Research Scientist, App and Ecosystem Trust

Google Google · Big Tech · Singapore

Research Scientist role focused on Android security and AI agent safety, developing and publishing novel research, designing neuro-symbolic and agent-powered systems for threat detection, and using machine learning for code analysis to combat malware and evasion techniques within the Android ecosystem.

What you'd actually do

  1. Identify problems and threat trends and support the team to move fast and come up with new ideas to manage the adversarial nature of the space and the adaptation of abusers, who would constantly come up with new strategies to bypass our protections.
  2. Develop and publish novel research on Android security and Artificial Intelligence (AI) agent safety, collaborating with global academic and technical partners to improve the defenses.
  3. Design neuro-symbolic and agent-powered systems to identify privacy threats, reverse-engineer apps, and ensure the safety of autonomous systems interacting with mobile web environments.
  4. Perform analysis on the codebases (e.g., Java, JavaScript, Native) using machine learning to detect malware, bypass obfuscation, and neutralize evasion techniques.
  5. Work with teams across Google such as AdSpam, Android Platform, Actor Trust, Google Research to address abuse problems together.

Skills

Required

  • PhD in Computer Science, a related field, or equivalent practical experience
  • 2 years of experience leading a research agenda
  • One or more scientific publication submission(s) for conferences, journals, or public repositories (such as CVPR, ICCV, NeurIPS, ICML, ICLR, etc.)

Nice to have

  • 2 years of coding experience
  • 2 years of experience leading research efforts and influencing other researchers

What the JD emphasized

  • novel research
  • AI agent safety
  • agent-powered systems
  • machine learning
  • detect malware
  • adversarial nature

Other signals

  • develop and publish novel research on Android security and AI agent safety
  • design neuro-symbolic and agent-powered systems
  • perform analysis on codebases using machine learning to detect malware