Remote AI Safety Specialist: Adversarial Testing & Failures

Mercor; Almada; Remote; EUR 33,000 - 61,000; Full time

Job summary

Mercor is seeking an AI Safety Specialist (Remote) to lead adversarial testing of conversational AI, annotate failures, and document vulnerabilities to strengthen client model safety. This contract role emphasizes strong communication and the ability to work asynchronously in a global, remote setting.

The ideal candidate has experience in AI adversarial work or cybersecurity, fluency in English and Portuguese, and the discipline to coordinate with teams across time zones while delivering

Qualifications

• Experience in AI adversarial testing or cybersecurity.
• Ability to work asynchronously in a global remote setting.
• Fluent in English and Portuguese.

Responsibilities

• Red-team conversational AI and annotate failures.
• Document vulnerabilities to improve model safety.
• Collaborate remotely with global teams to support clients.

Skills

AI adversarial work

Cybersecurity

Communication skills

Asynchronous work

Job description

Mercor is seeking an AI Safety Specialist (Remote) to lead adversarial testing of conversational AI, annotate failures, and document vulnerabilities to strengthen client model safety. This contract role emphasizes strong communication and the ability to work asynchronously in a global, remote setting.

The ideal candidate has experience in AI adversarial work or cybersecurity, fluency in English and Portuguese, and the discipline to coordinate with teams across time zones while delivering