About the Role
Join a project-based expert network supporting AI model evaluation and improvement. You will provide structured feedback that helps make model responses more accurate, helpful, and safe.
What You'll Do
- Rank and compare model responses to identify the strongest answer.
- Rate responses for accuracy, helpfulness, tone, and safety using detailed rubrics.
- Review individual reasoning steps and label them as correct or incorrect.
- Write concise rationales to explain ratings and decisions.
- Fact-check claims using reliable sources.
- Evaluate responses against content policies and style guides.
- Assess search result relevance and provide live feedback on model outputs during sessions.
What We're Looking For
- Zero to five years of relevant professional experience.
- Native or near-native written English proficiency and excellent attention to detail.
- Experience rating, reviewing, fact-checking, editing, moderating content, or grading work against written guidelines.
- Experience applying a rubric or style guide consistently across a large number of items.
- A degree or work background in a field where you can assess answers, such as writing, journalism, research, teaching, trust and safety, mathematics, science, law, medicine, or finance.
- Experience with AI data labeling, preference ranking, response rating, search quality rating, or ads rating is helpful but not required.
Compensation & Benefits
Compensation ranges from $20 to $45 per hour depending on the project, field, and experience level. You are paid for hours worked.
Location
Fully remote. Projects typically last from one week to several months.
Compensation Range: $41.6K - $93.6K
Originally posted on Himalayas
Time-zone restrictions (UTC offsets): -11, -10, -9.5, -9, -8, -7, -6, -5, -4, -3.5, -3, -2, -1, 0, 1, 2, 3, 3.5, 4, 4.5, 5, 5.5, 5.75, 6, 6.5, 7, 8, 8.75, 9, 9.5, 10, 10.5, 11, 12, 12.75, 13, 14