German Trust & Safety Data Trainer

German Trust & Safety Data Trainer

This remote, hourly‑paid contract role involves reviewing AI‑generated content and safety decisions, evaluating reasoning quality, and providing expert feedback to ensure that outputs are accurate, logical, and clearly explained. You will assess solutions for correctness and clarity, spot conceptual or methodological errors, fact‑check when needed, and rate/compare multiple responses based on safety and policy alignment. Fluency in German and the ability to make nuanced judgments across German and English content are required.

Exposure to sexual, violent, or psychologically disturbing material may occur as part of this work.

• Label and quality‑check safety data across categories such as hate/harassment, sexual content, self‑harm, violence, bias, illegal goods/services, malicious activities, malicious code, and deliberate misinformation.
• Perform red‑team and adversarial testing by identifying realistic attack patterns, edge cases, and policy gray areas; document rationales and recommend mitigations to reduce unsafe outcomes.
• Apply and localize safety policies consistently across German and English: detect cultural nuance, slang, coded language, and context shifts; elevate uncertainty using documented decision paths.

• Bachelor’s degree or higher in a relevant field (e.g., Communications, Linguistics, Psychology, Law/Policy, Security Studies) or equivalent professional experience.
• Near‑native or native German proficiency (reading/writing) for high‑precision safety labeling and cultural‑linguistic nuance.
• Minimum C1 English proficiency (reading/writing) for policy interpretation, prompt understanding, and consistent documentation.
• Experience in Trust & Safety, content moderation, policy operations, risk, compliance, investigations, or related safety functions (senior level).
• LLM red‑team / adversarial testing experience is required, including documented examples of edge‑case discovery and mitigation recommendations.
• Localization/translation experience is highly preferred; able to preserve meaning, severity, and intent across languages.
• Emotional resilience: comfortable annotating unsafe, explicit, or toxic content, including sexual, violent, or psychologically disturbing material.
• Highly detail‑oriented with strong judgment, consistency, and the ability to follow evolving written guidelines.
• Strong analytical writing: concise rationales, clear decision paths, and reproducible reasoning for disagreements.
• Secure and confidential handling of sensitive content; reliable remote work practices and time management.
• Strong hands‑on experience using tools such as Perplexity, Gemini, ChatGPT, and others.

Mid‑Senior level

Contract