Senior AI Red Team Operator - Adversarial Security Leader

Verizon; Irving (TX); Hybrid; USD 121,000 - 231,000; Full time

Benefits offered by this job

Health insurance

401(k) plan

Paid time off

Tuition assistance

Job summary

Verizon is seeking a Principal AI Red Team Operator to lead end-to-end adversarial assessments on AI/ML systems. You will mature the AI red team program with frameworks, playbooks, and automated tools, targeting LLMs, RAG architectures, and agents.

This role reports to the AI Security Operations Manager and requires collaboration with engineering and leadership. You will drive testing, reporting, and remediation guidance to strengthen the AI lifecycle while tracking evolving GenAI threats and

Qualifications

• Experience conducting AI evaluations or developing AI Safety/Security systems.
• Experience using adversarial prompting strategies to compromise enterprise AI systems.
• Experience using data poisoning, embedding inversion, or similar techniques.

Responsibilities

• Mature, scale and operationalize AI red team frameworks, playbooks and automated tools.
• Plan and conduct automated and manual adversarial assessments on enterprise AI systems.
• Present findings and risk insights to engineering teams and senior leadership with actionable remediation guidance.
• Research emerging GenAI threats and continuously enhance testing capabilities.

Skills

AI evaluations

Adversarial prompting

Data poisoning

Python

LLMs and AI security

Cloud platforms (GCP/AWS/Azure)

Independent work

Team collaboration

Education

Bachelor's degree or 4+ years experience

Job description

Verizon is seeking a Principal AI Red Team Operator to lead end-to-end adversarial assessments on AI/ML systems. You will mature the AI red team program with frameworks, playbooks, and automated tools, targeting LLMs, RAG architectures, and agents.

This role reports to the AI Security Operations Manager and requires collaboration with engineering and leadership. You will drive testing, reporting, and remediation guidance to strengthen the AI lifecycle while tracking evolving GenAI threats and