Senior AI Security Researcher

This is a fully remote position, open to applicants in North Carolina, +3 more states.

📋 Description

• Formulate and address open-ended AI security research inquiries that enable NVIDIA to understand, assess, and mitigate risks in frontier models, agentic systems, AI platforms, and AI-integrated products.

• Create practical methodologies, prototypes, evaluations, or tools that demonstrate how AI systems can fail under adversarial scenarios and identify strategies to reduce those risks.

• Investigate a variety of AI security challenges, including LLM and agent security, adversarial testing, model assessment, cyber-defense automation, vulnerability detection, secure deployment, and autonomous response.

• Convert research findings into actionable outcomes for engineering and security teams, featuring proof-of-concept demonstrations, benchmarks, technical guidance, mitigations, and secure-by-design recommendations.

• Collaborate with teams across offensive security, product security, AI research, platform, cloud, and infrastructure to align research insights with NVIDIA's most critical security objectives.

• Contribute to the development of NVIDIA's AI-security research strategy by mentoring peers, identifying emerging risks, and establishing repeatable practices for evaluating and safeguarding AI systems.

⛳️ Requirements

• Over 12 years of experience in AI security, cybersecurity research, applied machine learning research, offensive security, cyber defense, or related technical disciplines.

• Proven track record of original research and practical contributions, such as deployed security ML systems, AI-security evaluations, CVEs, patents, publications, conference presentations, open-source tools, production mitigations, or funded research initiatives.

• Proficient in constructing operational research systems using Python and modern ML/data tools like PyTorch, JAX, TensorFlow, scikit-learn, Pandas, NumPy, Spark, BigQuery, or similar platforms.

• Familiarity with one or more AI-security domains: LLM security, adversarial ML, model evaluation, agent security, prompt injection, model backdoors, data poisoning, model abuse, secure RAG, synthetic data, or AI-enabled security automation.

• Strong foundation in cybersecurity principles, encompassing threat modeling, adversary simulation, exploit or vulnerability research, malware analysis, network defense, threat hunting, detection engineering, digital forensics, secure code review, or incident-response automation.

• Capability to navigate ambiguous research challenges and real-world product constraints, translating results into prioritized recommendations and measurable security advancements.

• Bachelor's degree or equivalent experience in Computer Science, Machine Learning, Cybersecurity, or a related area.

• Experience leading AI-security research for significant models, AI platforms, security products, or large-scale production systems.

🏝️ Benefits

• Equity

• Benefits