AI Safety Experts — English & Finnish

• Red team conversational AI models and agents, including jailbreaks, prompt injections, misuse cases, bias exploitation, and multi-turn manipulation.
• Generate high-quality human data by annotating failures, classifying vulnerabilities, and flagging systemic risks.
• Apply structure to testing by following taxonomies, benchmarks, and playbooks to ensure consistency.
• Document findings reproducibly by producing reports, datasets, and attack cases for customer action.

• Prior red teaming experience, including AI adversarial work, cybersecurity, or socio-technical probing.
• Curiosity and an adversarial mindset, with the ability to instinctively push systems to their breaking points.
• A structured approach to testing, utilizing frameworks or benchmarks.
• Clear communication skills to explain risks to both technical and non-technical stakeholders.
• Adaptability to thrive when moving across different projects and customers.

• Specialization in Adversarial ML, including jailbreak datasets, prompt injection, RLHF/DPO attacks, and model extraction.
• Specialization in Cybersecurity, including penetration testing, exploit development, and reverse engineering.
• Specialization in Socio-technical risk, including harassment/disinformation probing, abuse analysis, and conversational AI testing.
• Skills in creative probing, such as psychology, acting, or writing for unconventional adversarial thinking.

RequiredAlso useful

Also listed: communication, adaptability.

Project and contracting process: Mercor. Applications continue on the provider's website.