AI Adversarial Specialist – Remote (English & Urdu)

About the role

Mercor is seeking an AI Adversarial Specialist to join its remote team. You will work on red‑teaming conversational AI models and agents, focusing on jailbreaks, prompt injections, and misuse scenarios while ensuring safety for English‑ and Urdu‑speaking users.

Key responsibilities

• Red‑team conversational AI systems to uncover jailbreaks, prompt injections, and other misuse cases.
• Generate high‑quality human data by annotating failures, classifying vulnerabilities, and flagging systemic risks.
• Follow established taxonomies, benchmarks, and playbooks to maintain consistent testing methodology.
• Document findings reproducibly, producing reports, datasets, and actionable attack cases for customers.
• Review AI outputs on sensitive topics such as bias and misinformation, with optional involvement in higher‑sensitivity projects.

Required profile

• Native fluency in English and Urdu.
• Prior experience in AI red‑teaming, cybersecurity, or socio‑technical risk analysis.
• Ability to communicate risks clearly to both technical and non‑technical stakeholders.

Required skills

• Red‑team testing of AI models.
• Adversarial machine‑learning concepts.
• Cybersecurity fundamentals.

What we offer

• Contract work with flexible hours.
• Competitive hourly rate of $20–$22.
• Fully remote setup allowing you to work from anywhere.