AI Red-Teamer — Adversarial AI Testing (Advanced); English & Arabic

AI/ML • Hourly Contract • Remote • $32.25/hour

Mercor is assembling an advanced red team to probe AI systems with adversarial inputs and generate safety data for customers.

This role requires native-level English & Arabic fluency and involves text-based evaluation of sensitive AI outputs.

What You Will Do

• Red-team conversational AI models with jailbreaks, prompt injection, and misuse cases.
• Annotate failures, classify vulnerabilities, and flag systemic risks.
• Apply taxonomies, benchmarks, and playbooks to keep testing consistent.
• Document reproducible reports and attack cases for customer use.

Who You Are

• Prior red teaming experience in AI, cybersecurity, or socio-technical probing.
• Native-level fluency in English and Arabic.
• Structured, framework-driven approach to adversarial testing.
• Clear communicator who can explain risks to technical and non-technical stakeholders.

More About the Opportunity

• Remote role with geography restricted to USA, Egypt, Saudi Arabia, and UAE.
• Contract work available full-time or part-time.
• Independent contractor engagement with weekly payments via Stripe or Wise.
• Participation in higher-sensitivity projects is optional and supported.

If you are a seasoned red teamer eager to make AI systems safer, apply today.