Senior Technical Program Manager - LLM Safety

At NVIDIA, we redefine what’s possible in digital imaging, personal computer gaming, and high-performance computing. Now, we lead the charge into AI’s unlimited potential. Our GPUs act as the brains of computers, robots, and self-driving cars. They enable these machines to understand and interact with the world in new ways. This is your chance to join a legacy of innovation and excellence. Work with the world’s best talent in a diverse and encouraging environment. Join us and make a lasting impact on the world!

This is an ambitious opportunity to work at the forefront of AI safety and contribute to groundbreaking advancements in technology.

What You’ll Be Doing:

• Lead cross-functional planning and execution for evaluation identification, selection and execution. With a focus on agentic-safety evaluations, including multi-turn tool calling, unsafe tool use, unintended actions, excessive autonomy, and multi-step failures.
• Track execution, results for all evals, analysis, and the mitigation and research plans for improvement that will stem from results analysis.  Translate evaluation findings into prioritized mitigation plans with accountable owners, committed dates, and measurable closure criteria
• Lead end-to-end safety planning and execution for Nemotron models, including program scope, dependencies, risks, resources, and release-readiness criteria.
• Translate technical safety priorities into executable programs covering LLM security, frontier risks, agentic safety, hallucinations
• Establish and track multi-turn, multi-modal, multi-lingual, long-context, and reasoning model evaluations across relevant use cases, domains, languages, modalities, and model releases.
• Establish governance for reviewing safety findings, assigning severity, determining release impact, and escalating unresolved risks. Ensure required safety evidence, approvals, exceptions, and risk-acceptance decisions are documented before release. Identify program gaps and critical dependencies early and drive decisions, escalations, recovery plans, and corrective actions.
• Build (with AI for Code/AI for Work) and maintain dashboards and executive reporting for evaluation coverage, critical findings, mitigation progress, residual risk, and release readiness.

…