Trust and Safety Data Trainer ( Multilingual) - Contract to Hire

Job Overview : We are hiring experienced Trust & Safety Data Trainers for a fully remote role. You will review AI-generated content and safety decisions, evaluate reasoning quality, and provide expert feedback to ensure outputs are accurate, safe, and aligned with policies. This role involves handling content that may be explicit, violent, or otherwise sensitive to help improve the world’s leading AI models. Candidates must be fluent in one of the following languages: Arabic, Spanish, French, Hebrew, Hindi, Japanese, Korean, Portuguese, or Chinese, with strong English proficiency. Key Responsibilities: . Label and quality-check AI content across safety categories: hate/harassment, sexual content, self-harm, violence, bias, misinformation, illegal activities, and malicious code . Perform red-teaming and adversarial testing to detect edge cases, policy gray areas, and unsafe outputs . Apply and localize safety policies with cultural, linguistic, and contextual accuracy . Document decisions, rationales, and mitigation recommendations clearly and reproducibly . Escalate uncertain or ambiguous cases following established guidelines. Requirements: . Native or near-native proficiency in one target language plus C1 English proficiency . Bachelor’s degree or equivalent professional experience in relevant fields (e.g., Communications, Linguistics, Psychology, Law/Policy, Security Studies) . Experience in Trust & Safety, content moderation, policy operations, compliance, investigations, or similar roles . Documented experience in red-teaming, adversarial testing, or AI safety evaluation . Emotional resilience to handle sensitive or disturbing content . Strong analytical, judgment, and written communication skills . Experience with AI tools and annotation platforms (e.g., ChatGPT, Gemini, Perplexity) Preferred Qualifications: . Localization/translation experience preserving meaning, intent, and severity across languages . Familiarity with structured guidelines, QA workflows, or content moderation frameworks Apply tot his job Apply To this Job