Remote AI Safety Red Team Specialist (EN/ID)

Obsidian; Singapore; On-site; SGD 153,000 - 230,000; Full time

Benefits offered by this job

Remote work options

Wellness resources

Job summary

Mercor is building a red team for AI safety, seeking human data experts to probe models with adversarial inputs and surface vulnerabilities. Work is text-based, with clear guidelines and wellness resources; participation in higher-sensitivity projects is optional.

You'll review AI outputs touching sensitive topics, generate high-quality data, annotate failures, and document reproducible results for customers. Remote work is supported across locations.

Qualifications

• Red team experience in AI adversarial work, cybersecurity or socio-technical probing.
• Curious and adversarial: pushes systems to breaking points.
• Structured: uses frameworks or benchmarks, not random hacks.
• Communicative: explains risks clearly to diverse stakeholders.
• Adaptable: able to move across projects and customers.

Responsibilities

• Red team conversational AI models and agents: jailbreaks, prompt injections, misuse cases, bias exploitation, multi-turn manipulation.
• Generate high-quality human data: annotate failures, classify vulnerabilities, and flag systemic risks.
• Apply structure: follow taxonomies, benchmarks, and playbooks to keep testing consistent.
• Document reproducibly: produce reports, datasets, and attack cases customers can act on.

Skills

Red team experience

Adversarial ML

Cybersecurity

Socio-technical risk

Creative probing

Tools

Penetration testing

Reverse engineering

Threat modeling

Job description

Mercor is building a red team for AI safety, seeking human data experts to probe models with adversarial inputs and surface vulnerabilities. Work is text-based, with clear guidelines and wellness resources; participation in higher-sensitivity projects is optional.

You'll review AI outputs touching sensitive topics, generate high-quality data, annotate failures, and document reproducible results for customers. Remote work is supported across locations.