Senior Research Scientist - AI Security

NVIDIA•Seattle, WA

10d•$168,000 - $304,750

About The Position

NVIDIA’s Agentic Cybersecurity Research team seeks an AI security researcher to innovate and improve protection of AI models, systems, and infrastructure. The role involves researching and developing technology that secures AI and systems through AI, including work on using models and agents to improve cybersecurity and securing models and agents. The team’s work contributes to advances in security and AI across the industry. The researcher will investigate new problems in AI and security, develop tools and processes to expose or remediate novel weaknesses in conventional and AI systems, and share research results through peer-reviewed publications or other channels with security, machine learning, and business communities. The role also involves sharing findings with teams across NVIDIA and applying research to relevant problems throughout the company.

Requirements

Master’s or PhD degree in Computer Science, Computer Engineering, Data Science, or a related field, or equivalent experience.
At least 5 years of experience in machine learning, cybersecurity, or closely related fields.
Demonstrated programming expertise.
Outstanding knowledge of AI and conventional software weaknesses and mitigation strategies.
In-depth understanding of Generative AI workflows.
Self-starter with a passion for growth, enthusiasm for continuous learning, and willingness to share findings.
Motivation and curiosity about new technologies.
Ability to present sophisticated technical information clearly.

Nice To Haves

Strong collaboration skills and a history of successful cross-functional projects.
Experience discovering CVEs or building successful tooling.
Knowledge of adversarial machine learning and techniques for defending against adversarial techniques.
Experience with secure software development practices.
Ability to build human-friendly resources.

Responsibilities

Conduct novel research to improve cybersecurity and AI, including using models and agents to improve cybersecurity and securing models and agents.
Contribute research on emerging problems in AI and security.
Develop tools and processes to expose or remediate novel weaknesses in conventional and AI systems.
Present and share research results through peer-reviewed venues and other channels with security, machine learning, and business communities.
Share findings with teams across NVIDIA and help solve relevant problems in different contexts.