AI Safety Red Team Expert: Behavioral Health & Model Eval

AuraOne; United States; Remote; USD 62.000 - 96.000; Vollzeit

Zusammenfassung

Ideal candidates have red-teaming or adversarial ML experience, strong written reporting skills, and comfort working in policy-grey areas with clear documentation. This is a remote, US-eligible contractor position.

Qualifikationen

• Red-teaming AI systems, security research, or adversarial ML experience.
• Strong written communication — reports become patch tickets.
• Comfort documenting what was attempted and why in policy-gray areas.

Aufgaben

• Design adversarial prompts that probe known weaknesses (jailbreak, policy bypass, prompt injection).
• Document every successful attack with reproduction steps and violated policy clause.
• Score model defenses across single-turn and multi-turn conversations.
• Triaged emerging attack vectors and routed them to the safety team with severity ratings.
• Maintain a personal library of attack patterns and propose new red-team rubrics.
• Calibrate against the broader red-team cohort to keep coverage and severity consistent.

Kenntnisse

Adversarial prompting

Red-team analysis

Policy taxonomy

Failure documentation

Behavioral Health AI Safety

Jobbeschreibung

Ideal candidates have red-teaming or adversarial ML experience, strong written reporting skills, and comfort working in policy-grey areas with clear documentation. This is a remote, US-eligible contractor position.