Manipulation Risk Red Team Specialist

About the role

Remote red-team track for stress-testing AI systems against adversarial prompts. Independent specialist contractor; hourly rate confirmed after interview. Reviewers craft attack scenarios, document failure modes and pair each jailbreak with the rubric clause violated. Nice to have: offensive security, AppSec, trust & safety background.

Requirements

• Demonstrated experience red-teaming AI systems, security research, or adversarial ML
• Strong written communication
• Comfort working in policy-grey areas with clear documentation
• Familiarity with prompt-injection, jailbreak, policy-bypass taxonomies
• Async availability at least 10 hours per week
• Authorized to work in the US

What you'll do

• Design adversarial prompts probing jailbreak, policy bypass, prompt injection
• Document successful attacks with reproduction steps
• Score model defenses across single and multi-turn conversations
• Triage emerging attack vectors and rate severity
• Maintain attack pattern library and propose red-team rubrics

Where this was posted

• Indeed

  last seen 1h ago
  View on Indeed