Remote AI Red Team & Safety Specialist

Mercor; Dublin; Remote; EUR 90,000 - 130,000; Full time

Job summary

Mercor is assembling a dedicated red team to probe AI models with adversarial inputs, surface vulnerabilities, and generate high-quality data to improve safety and trust for customers.

You will red-team conversational AI systems, apply structured frameworks, document reproducible attack cases, and collaborate with clients across projects to strengthen defenses while ensuring wellbeing guidelines and optional participation in higher-sensitivity tasks.

Qualifications

• Prior red-teaming experience in AI or cybersecurity.
• Curious and adversarial mindset that pushes systems to limits.
• Structured approach using frameworks or benchmarks.
• Able to explain risks to both technical and non-technical stakeholders.
• Adaptive to moving across projects and customers.

Responsibilities

• Red team conversational AI models: jailbreaks, prompt injections, misuse cases, bias exploitation.
• Generate high-quality human data: annotate failures, classify vulnerabilities, flag systemic risks.
• Document tests and results using structured reports and datasets.
• Follow taxonomies, benchmarks, and playbooks to ensure consistency.

Skills

Red team experience

Adversarial thinking

Structured approach

Clear communication

Adaptive across projects

Tools

Penetration testing

Exploit development

Reverse engineering

Job description

Mercor is assembling a dedicated red team to probe AI models with adversarial inputs, surface vulnerabilities, and generate high-quality data to improve safety and trust for customers.

You will red-team conversational AI systems, apply structured frameworks, document reproducible attack cases, and collaborate with clients across projects to strengthen defenses while ensuring wellbeing guidelines and optional participation in higher-sensitivity tasks.