Senior Machine Learning Engineer, AI Safety

Details

NVIDIA is seeking Deep Learning Scientists, AI Researchers, and Machine Learning Engineers to support its AI Safety and Responsibility efforts for Enterprise Risk Management. The role focuses on scaling safety for multimodal large language models (LLMs), including advanced agentic safety.

NVIDIA develops AI-based products across multiple domains and collaborates with leading AI companies as partners and customers. This position focuses on measuring and improving the security, content safety, and inclusivity of frontier models.

Core Focus Areas

• LLM Security: Backdoors, data poisoning, latent malicious behavior, and structural model vulnerabilities.
• Frontier Risks: Advanced alignment challenges, including model deception, manipulation, and loss-of-control scenarios.
• Agentic Safety: LLM-level safety for autonomous systems, including multi-turn tool calling, orchestration, and execution risks.
• Multi-Turn Safety Evaluation: Robust, scalable automated evaluation methodologies for conversational and iterative multi-turn use cases.

Responsibilities

• Develop datasets, specialized models, and algorithms to evaluate and benchmark models and end-to-end systems across LLM security, agentic safety, content safety, hallucinations, and machine learning fairness.
• Develop datasets and training recipes for model pre-training, mid-training, and post-training, including data-filtering components, reinforcement learning environments, and teacher models.
• Research and deploy model- and system-level techniques beyond post-training, such as instruction hierarchy and risk detection.
• Partner with engineers, data scientists, and research teams across NVIDIA to scale safety solutions.

Requirements

• Master’s or PhD in Computer Science, Electrical Engineering, or a related quantitative field, or equivalent experience.
• At least 8 years of proven experience in systems software engineering or machine learning engineering.
• At least 4 years of hands-on experience with post-training LLMs, including supervised fine-tuning (SFT), reinforcement learning from human or AI feedback (RLHF/RLAIF), safety data generation techniques, ablation studies, and production model deployment.
• At least 1 year of dedicated experience or research in LLM security, frontier risks, agentic safety, or multi-turn safety evaluation.
• In-depth knowledge of machine learning principles and frameworks, with PyTorch preferred.
• Strong Python programming skills.
• Experience working with large multimodal datasets and multimodal foundation models.
• Outstanding analytical problem-solving, collaboration, and communication skills.
• Behaviors that build trust, including humility, transparency, respect, and intellectual honesty.

Preferred Qualifications

• Primary-author publications on AI safety, alignment, or machine learning security at conferences such as NeurIPS, ICML, ICLR, or ACL.
• Contributions to open-source AI safety tools, benchmarks, datasets, or models.
• Experience with alignment or fine-tuning of vision-language models (VLMs) or any-to-text foundation models.

Compensation and Benefits

The base salary range is $184,000–$287,500 USD for Level 4 and $224,000–$356,500 USD for Level 5. Compensation is determined based on location, experience, and the pay of employees in similar positions. The role also includes eligibility for equity and benefits.

Applications will be accepted at least until September 4, 2026. This posting is for an existing vacancy. NVIDIA uses AI tools in its recruiting processes and is an equal opportunity employer.