Senior AI Safety Engineer - Guardrails & Equity

NVIDIA Corporation; United States; Remote; USD 184,000 - 357,000; Full time

Job summary

NVIDIA is seeking an AI Engineer for its AI Safety Engineering team. You will architect agentic automation loops and develop safety models to detect harmful content and enforce run-time guardrails.

You’ll guide automated research frameworks using LLMs to discover vulnerabilities and produce safe, scalable production systems. Ideal candidates have 8+ years in ML or software engineering, strong Rust/Python skills, PyTorch experience, and a proven ability to work across research and platform teams

Qualifications

• Education: Bachelor’s or Master’s in Computer Science, ML, Software Engineering, or equivalent.
• 8+ years in ML engineering or software engineering.
• Proven experience with LLMs, multi-turn tool-calling, orchestration, and autonomous agent loops.
• Strong programming in Rust or Python; PyTorch experience required.
• Problem solving in ambiguous, open-ended AI safety contexts.

Responsibilities

• Architect agentic automation loops spanning safety evaluations, red-teaming, and model testing.
• Develop safety models for harmful content detection and run-time guardrails.
• Build automated research frameworks using LLMs for discovery and data generation.
• Scale proof-of-concept safety ideas into robust production solutions.
• Collaborate with researchers, platform engineers, and data scientists to bridge theory and production.

Skills

Rust

Python

PyTorch

LLMs

Agentic automation

Education

Bachelor's or Master's in CS/ML/SE

Tools

LangChain

AutoGPT

Semantic Kernel

Job description

NVIDIA is seeking an AI Engineer for its AI Safety Engineering team. You will architect agentic automation loops and develop safety models to detect harmful content and enforce run-time guardrails.

You’ll guide automated research frameworks using LLMs to discover vulnerabilities and produce safe, scalable production systems. Ideal candidates have 8+ years in ML or software engineering, strong Rust/Python skills, PyTorch experience, and a proven ability to work across research and platform teams