Remote AI Safety Red Team Specialist

Mercor; Sydney; Remote; AUD 90,000 - 130,000; Full time

Job summary

Mercor invites a human data expert to remotely probe AI models with adversarial inputs, surface vulnerabilities, and generate red team data that enhances AI safety. You will annotate failures, classify vulnerabilities, and follow playbooks to keep testing consistent.

Clear documentation and reproducible attack cases will be produced for customer use. Native English and Odia fluency are required, with the ability to explain reasoning to diverse audiences.

Qualifications

• Native fluency in English and Odia is required.
• You bring strong judgment about language and content and can explain why an AI response is accurate, complete, and appropriate.
• You are rigorous and notice subtle errors, inconsistencies, and gaps that others might miss.
• You’re structured and follow guidelines and quality standards consistently.

Responsibilities

• Red team conversational AI models and agents: jailbreaks, prompt injections, misuse cases, bias exploitation, multi-turn manipulation
• Generate high-quality human data: annotate failures, classify vulnerabilities, and flag systemic risks
• Apply structure: follow taxonomies, benchmarks, and playbooks to keep testing consistent
• Document reproducibly: produce reports, datasets, and attack cases customers can act on

Skills

English & Odia

Judgment

Rigorous

Structured

Communicative

Adaptable

Job description

Mercor invites a human data expert to remotely probe AI models with adversarial inputs, surface vulnerabilities, and generate red team data that enhances AI safety. You will annotate failures, classify vulnerabilities, and follow playbooks to keep testing consistent.

Clear documentation and reproducible attack cases will be produced for customer use. Native English and Odia fluency are required, with the ability to explain reasoning to diverse audiences.