Remote AI Safety Red Team Specialist

Mercor; Toronto; Remote; CAD 85,000 - 125,000; Full time

Job summary

Mercor is building a remote red team to stress test AI systems. You will engage in adversarial testing, exploring biases, safety gaps, and guardrail weaknesses.

The role emphasizes rigorous methodology, clear reporting, and collaboration with teams across clients. You’ll produce actionable datasets and case reports to help improve safety.

You should have hands-on red-teaming, cybersecurity, and risk-analysis experience, plus strong communication skills and the ability to move across projects.

Qualifications

• You have prior red teaming experience focusing on AI adversarial tasks.
• You can apply frameworks or benchmarks to testing scenarios.
• You communicate risks clearly to technical and non-technical stakeholders.

Responsibilities

• Red team conversational AI models and agents with jailbreaks and prompt injections.
• Generate reproducible artifacts: reports, datasets, attack cases.
• Follow taxonomies and playbooks to keep testing consistent.
• Document findings for customers to act on.

Skills

Red team experience in AI adversarial

Adversarial mindset

Cybersecurity knowledge

Socio-technical risk assessment

Structured thinker

Clear communicator

Adaptability across projects

Curiosity and persistence

Job description

Mercor is building a remote red team to stress test AI systems. You will engage in adversarial testing, exploring biases, safety gaps, and guardrail weaknesses.

The role emphasizes rigorous methodology, clear reporting, and collaboration with teams across clients. You’ll produce actionable datasets and case reports to help improve safety.

You should have hands-on red-teaming, cybersecurity, and risk-analysis experience, plus strong communication skills and the ability to move across projects.