Remote AI Safety Red Team Engineer

Obsidian; Singapore; Remote; SGD 89,000 - 127,000; Full time

Job summary

Mercor is assembling a remote red team to probe AI models with adversarial inputs, surface vulnerabilities, and generate data that makes AI safer for customers. The work involves reviewing AI outputs on sensitive topics with clear guidelines.

You’ll document reproducible reports and contribute to safer AI systems across projects and customers. You should have prior red-teaming experience and be comfortable moving across diverse tasks, while articulating risks to both technical and non-technical

Qualifications

• Prior red teaming experience in AI adversarial work, cybersecurity, or socio-technical probing.
• Curious and adversarial, able to push systems to breaking points.
• Structured thinker who uses benchmarks or playbooks.
• Clear communicator for technical and non-technical stakeholders.
• Adaptable 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 risks.
• Apply structure: use taxonomies, benchmarks, and playbooks for testing consistency.
• Document reproducibly: produce reports, datasets, and attack cases for customers.

Skills

AI red teaming

Cybersecurity

Socio-technical probing

Adversarial mindset

Structured frameworks

Clear communicator

Adaptable

Job description

Mercor is assembling a remote red team to probe AI models with adversarial inputs, surface vulnerabilities, and generate data that makes AI safer for customers. The work involves reviewing AI outputs on sensitive topics with clear guidelines.

You’ll document reproducible reports and contribute to safer AI systems across projects and customers. You should have prior red-teaming experience and be comfortable moving across diverse tasks, while articulating risks to both technical and non-technical