Hebrew/English Bilingual Expert

SME Careers; United States; Remote; USD 3,444,000 - 5,510,000; Part time

Benefits offered by this job

Fully remote

Flexible hours

Job summary

A leading AI Data Services company is seeking a bilingual content evaluator to review AI-generated responses and create training content. The ideal candidate will have native or near-native Hebrew proficiency, minimum C1 English proficiency, and a bachelor's degree in a related field. Responsibilities include evaluating AI responses, ensuring model integrity, and optimizing AI performance. This fully remote role offers flexible hours and requires at least 3 years of relevant experience.

Qualifications

• Bachelor’s degree (or higher) in a relevant field.
• 3+ years of professional experience in relevant fields.
• Excellent writing and editing skills in Hebrew.
• 3+ years in translation, localization, editorial QA, content quality, or linguistic review.

Responsibilities

• Create detailed prompts and responses for AI training.
• Evaluate and rank AI responses to enhance model accuracy.
• Test AI models for biases and inaccuracies.

Skills

Native or near-native Hebrew proficiency

C1 English proficiency

Attention to detail

Fact-checking skills

Ability to identify reasoning gaps

Experience with AI data training

Education

Bachelor’s degree in Linguistics, Translation, or related field

Job description

You will review AI-generated Hebrew and English responses and/or generate high-quality bilingual training content, evaluating reasoning quality and step-by-step problem‑solving while providing expert feedback that helps models produce answers that are accurate, logical, and clearly explained. You will assess solutions for accuracy, clarity, and adherence to the prompt; identify errors in methodology or conceptual understanding; fact‑check information; write high‑quality explanations and model solutions that demonstrate correct methods; and rate and compare multiple AI responses based on correctness and reasoning quality. This is an hourly paid, fully remote contractor role with flexible work‑from‑anywhere hours. This job is with SME Careers (a subsidiary of SuperAnnotate), a fast‑growing AI Data Services company providing training data for many of the world’s largest AI companies and foundation model labs—your work directly helps improve the world’s premier AI models.

Key Responsibilities

• Develop AI Training Content: Create detailed prompts in various topics and responses to guide AI learning, ensuring the models reflect a comprehensive understanding of diverse subjects (Hebrew/English; requires C1+ English).
• Optimize AI Performance: Evaluate and rank AI responses to enhance the model’s accuracy, fluency, and contextual relevance (Hebrew/English; requires C1+ English).
• Ensure Model Integrity: Test AI models for potential inaccuracies or biases, validating their reliability across use cases (Hebrew/English; requires C1+ English).

Your Profile

• Bachelor’s degree (or higher) in Linguistics, Translation, Hebrew Language, Communications, Journalism, or a related field.
• Native or near‑native Hebrew proficiency with strong writing and editing skills across formal and informal registers.
• Minimum C1 English proficiency (reading and writing) for bilingual evaluation and instruction adherence.
• 3+ years of professional experience in translation, localization, editorial QA, content quality, or linguistic review (or equivalent).
• Excellent attention to detail and ability to apply detailed rubrics consistently in high‑volume, hourly contractor work.
• Strong fact‑checking habits and comfort validating claims using reputable sources when required by task guidelines.
• Ability to identify reasoning gaps, methodological errors, and unclear explanations—even when language is fluent.
• Comfort working independently in a remote, asynchronous environment with reliable availability and communication.
• Prior experience with AI data training, annotation, or evaluation workflows is strongly preferred.
• Familiarity with Israeli cultural context and terminology norms across common domains (news, education, consumer, tech) is preferred.