AI Security Red Team Specialist

OpenAI; San Francisco (CA); Hybrid; USD 190,000 - 230,000; Full time

Benefits offered by this job

Relocation assistance

Hybrid work model (3 days in office)

Job summary

OpenAI is seeking a Red Team Specialist focused on cyber to evaluate model cyber capabilities and safeguards, assessing what attackers could leverage and whether safeguards stay effective against sophisticated techniques.

The role blends scaled evaluations with hands-on testing, allowing deep expertise in cybersecurity or model evaluation and encouraging collaboration across policy, engineering, and product teams. Relocation assistance is provided for this hybrid role in SF or Seattle.

Qualifications

• Experience in cybersecurity, model evaluation, or both, with ability to write code and build testing tools.

Responsibilities

• Design and run evaluations of model cyber capabilities and safeguards, including policy adherence and jailbreaking defenses.
• Conduct hands-on testing to understand model capabilities used by security practitioners.
• Build and improve automated testing infrastructure for repeatable measurement and analysis.
• Test novel abuse risks in agentic systems and translate findings into actionable risk assessments.
• Contribute time to Safety Bug Bounty work where cyber expertise is required.
• Communicate findings effectively to Security, Research, Product, Policy, and Engineering teams.

Skills

Cybersecurity

Penetration testing

Vulnerability research

Red-team operations

AI model evaluation

Automated testing

Job description

OpenAI is seeking a Red Team Specialist focused on cyber to evaluate model cyber capabilities and safeguards, assessing what attackers could leverage and whether safeguards stay effective against sophisticated techniques.

The role blends scaled evaluations with hands-on testing, allowing deep expertise in cybersecurity or model evaluation and encouraging collaboration across policy, engineering, and product teams. Relocation assistance is provided for this hybrid role in SF or Seattle.