AI Safety Engineer

This contract role focuses on AI Safety for Large Language Models (LLMs) at Nvidia, working with research and engineering teams. You will develop datasets, moderator models, and implement techniques to improve the safety, fairness, and inclusivity of LLMs, including multi-lingual and multi-modal models.

• Develop datasets and moderator models for LLM evaluation
• Implement techniques for bias detection and mitigation in LLMs
• Define and track metrics for responsible LLM behavior
• Contribute to safety tools and repositories
• Proficient in Python, ML frameworks, and data manipulation
• Experience with GenAI security and content safety areas

• No visa sponsorship is available.