AI Safety Expert - Red Team

Mercor; Johannesburg; À distance; ZAR 1 322 000 - 1 818 000; Plein temps

Résumé du poste

Mercor is building a remote red-team for AI safety projects focused on adversarial inputs, content review, and robust evaluation. You will probe conversational AI models, surface vulnerabilities, and generate actionable data for customers.

The role emphasizes curiosity, systematic testing, and clear communication of risks to both technical and non-technical stakeholders, with flexibility to work across multiple projects and clients.

Qualifications

• Red team background in AI adversarial work, cybersecurity, or socio-technical probing.
• Curious and adversarial mindset with tendency to push systems to limits.
• Structured approach using frameworks or benchmarks, not random hacks.
• Excellent communication to explain risks to technical and non-technical stakeholders.
• Able to move across projects and customers with ease.

Responsabilités

• Red team conversational AI models: jailbreaks, prompt injections, bias testing.
• Generate high-quality human data: annotate failures and flag vulnerabilities.
• Apply structure: follow taxonomies, benchmarks, and playbooks for consistency.
• Document reproducibly: produce reports, datasets, and attack cases for customers.

Connaissances

Red team experience

Adversarial thinking

Structured testing

Clear communication

Adaptability

Description du poste

Location: Remote

Fluent Language Skills Required: English & Portuguese (global, excluding Brazilian Portuguese). Native fluency in English and Portuguese (global, excluding Brazilian Portuguese) is required for this position.

Why This Role Exists

At Mercor, we believe the safest AI is the one that is already been attacked - by us. We are assembling a red team for this project - human data experts who probe AI models with adversarial inputs, surface vulnerabilities, and generate the red team data that makes AI safer for our customers.

This project involves reviewing AI outputs that touch on sensitive topics such as bias, misinformation, or harmful behaviors. All work is text-based, and participation in higher-sensitivity projects is optional and supported by clear guidelines and wellness resources. Before being exposed to any content, the topics will be clearly communicated.

What You'll Do

• 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

Who You Are

• You bring prior red teaming experience (AI adversarial work, cybersecurity, socio-technical probing)

• You're curious and adversarial: you instinctively push systems to breaking points

• You're structured: you use frameworks or benchmarks, not just random hacks

• You're communicative: you explain risks clearly to technical and non-technical stakeholders

• You're adaptable: thrive on moving across projects and customers

Nice-to-Have Specialties

• Adversarial ML: jailbreak datasets, prompt injection, RLHF/DPO attacks, model extraction

• Cybersecurity: penetration testing, exploit development, reverse engineering

• Socio-technical risk: harassment/disinfo probing, abuse analysis, conversational AI testing

• Creative probing: psychology, acting, writing for unconventional adversarial thinking

What Success Looks Like

• You uncover vulnerabilities automated tests miss

• You deliver reproducible artifacts that strengthen customer AI systems

• Evaluation coverage expands: more scenarios tested, fewer surprises in production

• Mercor customers trust the safety of their AI because you've already probed it like an adversary

Why Join Mercor

• Build experience in human data-driven AI red teaming at the frontier of safety

• Play a direct role in making AI systems more robust, safe, and trustworthy