AI Safety Expert at Mercor

AI Safety Expert | $85–185/hr | Worldwide Remote

Join Mercor's specialized safety team to assess and stress-test frontier AI models, helping build more robust and trustworthy AI systems through adversarial testing and vulnerability analysis.

Why This Role Exists

At Mercor, we believe the safest AI is the one that's already been attacked — by us. We're building a team of safety specialists to rigorously assess the capabilities and limitations of frontier models, ensuring they're secure before deployment.

What You'll Do

• Generate high-quality human data by annotating failures, classifying vulnerabilities, and flagging systemic risks in AI systems
• Apply structured methodologies using established taxonomies, benchmarks, and playbooks to maintain consistent testing standards
• Document findings reproducibly through detailed reports, datasets, and attack cases that customers can immediately act upon
• Work flexibly across diverse projects, from LLM jailbreak testing to socio-technical abuse scenario analysis

Who You Are

• Naturally curious and adversarial, with an instinct for pushing systems to their breaking points
• Structured in your approach, leveraging frameworks and benchmarks rather than random testing
• Excellent communicator who can explain complex risks clearly to both technical and non-technical stakeholders
• Highly adaptable and thrive when moving between different projects and customer engagements

Why Join Mercor

• Build valuable experience in human data and AI safety work at the cutting edge of the field
• Play a direct role in making AI systems more robust, safe, and trustworthy for real-world deployment
• Work with leading AI labs and enterprises on frontier model development

About Mercor

Mercor partners with leading AI labs and enterprises to train frontier models using human expertise. Our work focuses on enhancing AI systems through rigorous testing and safety protocols.

We consider all qualified applicants without regard to legally protected characteristics and provide reasonable accommodations upon request.