Hebrew AI Data Trainer - Remote

Hebrew AI Data Trainer - Remote Location: Remote This is a fully remote, hourly‑paid contractor role focused on improving the quality, reasoning ability, and linguistic accuracy of AI systems. You will evaluate AI‑generated responses in Hebrew and English, assess step‑by‑step reasoning, and provide structured expert feedback to improve clarity, correctness, and alignment with user intent. Your work directly supports leading AI research organizations by helping refine model performance across multilingual and reasoning‑intensive tasks. Key Responsibilities Review and evaluate AI‑generated responses in Hebrew and English for accuracy, clarity, and logical reasoning Assess step‑by‑step problem‑solving quality and identify conceptual or methodological errors Rate and compare multiple AI outputs based on correctness, fluency, and adherence to prompts Write high‑quality model answers that demonstrate correct reasoning and clear explanation Develop bilingual AI training content (Hebrew/English) across diverse subject areas Identify inconsistencies, hallucinations, or weak reasoning in model outputs Apply structured evaluation rubrics consistently in high‑volume annotation workflows Fact‑check information using reliable sources when required Ensure model outputs reflect proper linguistic, cultural, and contextual accuracy Required Qualifications Bachelor’s degree (or higher) in Linguistics, Translation, Hebrew Language, Communications, Journalism, or a related field Native or near‑native Hebrew proficiency, including formal and informal registers Minimum C1 English proficiency (reading and writing) At least 3+ years of experience in translation, localisation, editorial QA, linguistic review, or content quality roles Strong ability to detect reasoning errors, unclear explanations, and logical inconsistencies Excellent attention to detail and ability to follow structured evaluation guidelines Strong fact‑checking skills using reliable sources when needed Key Competencies Analytical thinking and structured reasoning evaluation High‑level bilingual writing and editing (Hebrew/English) Consistency in applying evaluation rubrics Strong editorial judgment and clarity in feedback Ability to work efficiently in high‑volume task environments Preferred Qualifications Experience in AI training data, LLM evaluation, or prompt engineering Background in journalism, editorial QA, or linguistic research Familiarity with AI tools such as ChatGPT or similar systems Experience working with structured annotation or QA platforms #J-18808-Ljbffr