Ai Safety Red Teamer Expert

Mercor; Viseu; Sur place; EUR 70 000 - 100 000; Plein temps

Résumé du poste

Mercor is seeking an experienced professional to advance frontier AI safety through adversarial testing and red-teaming initiatives in Portugal. You will design challenging prompts, uncover model weaknesses, and assess AI behavior across high-risk, ambiguous topics, contributing to safety benchmarks and cross-disciplinary collaboration with AI researchers.

The role emphasizes rigorous analysis, documentation of vulnerabilities, and proactive engagement with safety teams to strengthen model

Qualifications

• Must have strong analytical reasoning and prompt design abilities.
• Excellent written communication skills.
• Experience designing adversarial prompts or evaluating frontier AI systems.

Responsabilités

• Identify vulnerabilities in frontier AI systems through adversarial testing.
• Design adversarial prompts to stress-test frontier AI models.
• Identify jailbreaks, unsafe behaviours, hallucinations, and policy failures.
• Evaluate model robustness across misinformation, cyber, biosecurity, fraud, political content, and other sensitive domains.
• Document vulnerabilities and contribute to safety benchmarking and red-teaming reports.
• Collaborate with AI researchers to improve model alignment, robustness, and safety.

Connaissances

Adversarial prompt design

Analytical reasoning

Written communication

Frontier AI evaluation

Formation

Bachelor's degree or higher in CS/Cybersecurity/Journalism/Communications/Policy

Description du poste

Responsibilities

• to identify vulnerabilities in frontier AI systems through adversarial testing. You will design challenging prompts, uncover model weaknesses, and evaluate AI behavior across complex, high-risk, and ambiguous ("grey-area") topics.
• Design adversarial prompts to stress-test frontier AI models.
• Identify jailbreaks, unsafe behaviours, hallucinations, and policy failures.
• Evaluate model robustness across misinformation, cyber, biosecurity, fraud, political content, and other sensitive domains.
• Document vulnerabilities and contribute to safety benchmarking and red-teaming reports.
• Collaborate with AI researchers to improve model alignment, robustness, and safety.

Required Qualifications

• Bachelor's degree or higher in Computer Science, Cybersecurity, Journalism, Communications, Psychology, Biology, Chemistry, Public Policy, or a related discipline.
• 5+ years of professional experience in AI Safety, AI Red Teaming, Trust & Safety, cybersecurity, investigative journalism, life sciences, or a related field.
• Strong analytical reasoning, prompt design, and written communication skills.
• Experience designing adversarial prompts or evaluating frontier AI systems.

Preferred Qualifications

• Experience with AI Red Teaming, RLHF, SFT, AI Alignment, or Trust & Safety.
• Familiarity with jailbreak testing, prompt engineering, or adversarial evaluation methodologies.
• Expertise in one or more grey-area domains, including cyber, biosecurity, political content, misinformation, or scientific safety.

Why Join?

• Help secure and strengthen the next generation of frontier AI models.
• Work on cutting-edge adversarial testing alongside leading AI researchers and safety teams.
• Influence how AI systems respond to complex, real-world safety challenges.