German AI Data Trainer – Remote

About the role

This remote, hourly‑paid contractor position focuses on improving the safety, reliability, and reasoning quality of AI systems. You will evaluate AI‑generated content in both German and English, applying linguistic and cultural judgment to ensure outputs are accurate, logically sound, and compliant with safety policies.

Key responsibilities

• Review AI‑generated responses for accuracy, clarity, reasoning quality, and safety compliance.
• Identify logical, methodological, or conceptual errors in model outputs.
• Rate and compare multiple AI responses based on correctness, safety alignment, and policy adherence.
• Conduct fact‑checking and validation where required.
• Apply safety policies consistently across German and English content.
• Perform red‑teaming and adversarial testing to uncover vulnerabilities and unsafe behaviors.
• Document findings with clear, structured reasoning and reproducible decision paths.
• Detect cultural nuance, slang, coded language, and contextual shifts in both languages.
• Recommend improvements to safety guidelines and mitigation strategies.

Required profile

• Bachelor’s degree or higher in Linguistics, Communications, Psychology, Law, Policy, Security Studies, or related field (or equivalent experience).
• Native or near‑native German proficiency (reading and writing) and minimum C1 English proficiency.
• Experience in Trust & Safety, content moderation, policy enforcement, compliance, investigations, or risk operations.
• Proven experience in LLM red‑teaming or adversarial testing.
• Strong analytical ability to localize content across languages while preserving intent and severity.
• Excellent written communication with clear, structured reasoning.
• High attention to detail and consistency in applying policy guidelines.
• Comfortable working independently in a remote contractor environment.

Required skills

• AI data annotation and model evaluation.
• Familiarity with AI tools such as ChatGPT, Gemini, Perplexity.
• LLM red‑teaming and adversarial testing.
• Structured safety frameworks or taxonomy‑based labeling systems.