AI-Driven Offensive Security Engineer — Red Team & Exploits

NVIDIA Corporation; Yorktown (VA); On-site; USD 180,000 - 240,000; Full time

Benefits offered by this job

Equity compensation

Benefits package

Job summary

NVIDIA Corporation in Virginia's Product Security organization seeks a Senior Offensive Security Engineer to push the frontier of AI-driven offensive security. You will craft autonomous red team agents, perform reconnaissance, hypothesis-driven exploitation, and evidence capture against NVIDIA-owned targets.

We expect 12+ years in security engineering with at least 4 years in offensive security, strong Python production code, and hands-on Exploit development across web, cloud/Kubernetes, or

Qualifications

• 12+ years in security engineering, with at least 4 years in offensive security.
• Hands-on exploitation in web, cloud/Kubernetes, or systems/binary.
• Strong Python software engineering; production code.
• Experience building with LLMs: agent frameworks and multi-agent orchestration.
• Ethical and responsible approach to offensive testing.

Responsibilities

• Develop autonomous and semi-autonomous LLM agents for reconnaissance and exploitation.
• Craft new attack capabilities and target classes for AI-driven security.
• Build verification harnesses to reduce false positives.
• Operate on NVIDIA-owned targets with safe autonomy.
• Contribute to AI security research.

Skills

Security engineering

Offensive security

Python

LLM-based tooling

Red teaming

Education

Bachelor's degree

Tools

Kubernetes

OpenShift

GitOps

Job description

NVIDIA Corporation in Virginia's Product Security organization seeks a Senior Offensive Security Engineer to push the frontier of AI-driven offensive security. You will craft autonomous red team agents, perform reconnaissance, hypothesis-driven exploitation, and evidence capture against NVIDIA-owned targets.

We expect 12+ years in security engineering with at least 4 years in offensive security, strong Python production code, and hands-on Exploit development across web, cloud/Kubernetes, or