AI Safety Expert - Red Team

About the role

We are looking for an AI Safety Expert to join our Red Team on a contract basis. You will work remotely to probe conversational AI models and agents, uncovering jailbreaks, prompt injections, misuse scenarios, and bias exploitation.

Key responsibilities

• Red‑team conversational AI systems to identify and document jailbreaks, prompt injections, misuse cases, and bias exploitation.
• Generate high‑quality human data by annotating model failures, classifying vulnerabilities, and flagging systemic risks.
• Apply structured testing approaches using taxonomies, benchmarks, and playbooks to ensure consistent and repeatable assessments.
• Document findings reproducibly, producing reports, datasets, and actionable attack cases for customers.
• Work independently and asynchronously, meeting deadlines while contributing to improvements in AI model performance.

Required profile

• Fluent in English and Finnish.
• Prior experience in red‑team operations, AI adversarial research, cybersecurity, or socio‑technical probing.
• Ability to communicate risks clearly to both technical and non‑technical stakeholders.

Required skills

• Adversarial Machine Learning (jailbreak datasets, prompt injection, model extraction).
• Penetration testing and exploit development.
• Socio‑technical risk analysis, including harassment, disinformation probing, and abuse analysis.