Remote AI Safety Red Team Specialist

Obsidian; Greater London; Remote; GBP 60,000 - 120,000; Full time

Job summary

Mercor is assembling a red team to probe AI models with adversarial inputs. The role involves reviewing AI outputs on sensitive topics, with topics clearly communicated before exposure.

You will engage in high-safety testing across projects, contributing data and reports to strengthen clients' AI safety postures. This position emphasizes structured methodologies, reproducible artifacts, and collaboration with diverse teams, while offering wellness resources and clear guidelines for

Qualifications

• Previous red teaming experience in AI adversarial work or cybersecurity.
• Curious and adversarial mindset with a drive to probe systems.
• Structured approach using benchmarks or frameworks.
• Clear communication of risks to technical and non-technical audiences.
• Adaptable across different 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.
• Apply structure: follow taxonomies, benchmarks, and playbooks to keep testing consistent.
• Document reproducibly: produce reports, datasets, and attack cases for customers.

Skills

Red Teaming

Cybersecurity

AI Adversarial Testing

Documentation

Communication

Tools

AI Red Team Tools

Job description

Mercor is assembling a red team to probe AI models with adversarial inputs. The role involves reviewing AI outputs on sensitive topics, with topics clearly communicated before exposure.

You will engage in high-safety testing across projects, contributing data and reports to strengthen clients' AI safety postures. This position emphasizes structured methodologies, reproducible artifacts, and collaboration with diverse teams, while offering wellness resources and clear guidelines for