Senior AI Safety Engineer – LLM Security & Alignment

NVIDIA; Santa Clara (CA); On-site; USD 200,000 - 288,000; Full time

Benefits offered by this job

Equity

Benefits

Job summary

NVIDIA is seeking talented Deep Learning Scientists / AI Researchers / Machine Learning Engineers to join our rapidly growing AI Safety and Responsibility efforts for Enterprise Risk Management. You will tackle innovative problems in machine learning, focusing on scaling safety for multi-modal LLMs, including advanced agentic safety.

We collaborate with leading AI partners and customers to improve security, content safety, and inclusivity of frontier models.

Qualifications

• Master’s or PhD in Computer Science, Electrical Engineering, or a related quantitative field (or equivalent experience).
• 8+ years of proven experience in systems software engineering or machine learning engineering.
• Post-Training Experience: 4+ years of hands-on work in post-training of LLMs, including SFT, RLHF/RLAIF, safety data generation, ablation studies, production deployment.
• Core Safety Expertise: 1+ years in at least one of: LLM Security, Frontier Risks, Agentic Safety, Multi-turn Safety Evaluation.

Responsibilities

• Evaluation: Develop datasets and models to evaluate/benchmark LLMs and end-to-end systems across safety tracks.
• Model Pre-Training, Mid-Training, Post-Training: Create datasets and recipes for data filtering and training data pipelines.
• Model & system level techniques beyond post-training: Research and deploy new approaches like Instruction Hierarchy or Risk Detection.
• Cross-Functional Collaboration: Work with engineers, data scientists, and research teams to scale safety solutions.

Skills

Python

PyTorch

Machine Learning

Multimodal

Education

Master’s or PhD in CS/EE

Tools

RLHF/RLAIF

Job description

NVIDIA is seeking talented Deep Learning Scientists / AI Researchers / Machine Learning Engineers to join our rapidly growing AI Safety and Responsibility efforts for Enterprise Risk Management. You will tackle innovative problems in machine learning, focusing on scaling safety for multi-modal LLMs, including advanced agentic safety.

We collaborate with leading AI partners and customers to improve security, content safety, and inclusivity of frontier models.