Job Description
This position develops technologies in privacy-preserving systems for agents using machine learning and large language models. Researchers conduct experiments and publish findings in areas like CVPR or NeurIPS, focusing on security and privacy. The work shapes systems that protect user data in AI interactions.
Google conducts fundamental research in computer science and product innovation. The company addresses challenges in hardware, software, and natural language processing. It shares discoveries through open source and journal publications.
Researchers author papers and build prototypes using Python for simulations in differential privacy and contextual integrity. They collaborate with internal teams and universities on security technologies. Work includes simulations, data processing, and hypothesis refinement. One constraint is owning research agendas in complex, ambiguous domains.
The base salary ranges from $141,000 to $202,000 plus bonus, equity, and benefits. The role offers a full-time schedule in unspecified locations.
Responsibilities
- Develop personal agents that transform computing for people around the world by conducting research and innovating in privacy-preserving technologies and security protections for agentic systems
- Collaborate closely with researchers and with the many partner and customer teams across Google
- Author research papers to share and generate impact of research results across the team and in the research community
- Develop research prototypes or conduct simulations to further evaluate the impact of research, finalize hypotheses, and refine the research methodology under minimal guidance
Requirements
- PhD degree in Computer Science, a related field, or equivalent practical experience
- One or more scientific publication submission(s) for conferences, journals, or public repositories (such as CVPR, ICCV, NeurIPS, ICML, ICLR, etc.) in the fields of security and privacy, systems, large language models, or related areas
- 2 years of coding experience preferred
- 1 year of experience owning and initiating research agendas preferred
- Research experience comparable to a postdoc at the intersection of security and privacy, systems, large language models, or closely related areas preferred
- Domain expertise in machine learning security & privacy as well as demonstrated interest and proficiency in the core concepts of privacy-preserving technologies (such as differential privacy or contextual integrity) preferred