Clera
Join a remote expert network as a senior rater and reviewer, evaluating complex AI model outputs and helping calibrate other graders. Your work supports the development of high-quality, trustworthy AI systems across specialized fields.
Rank and rate expert-level model answers using detailed rubrics and write clear rationales.
Evaluate steps in model reasoning across fields such as mathematics, coding, law, medicine, and finance.
Fact-check complex claims against primary sources and verify accuracy.
Assess answers against content policies, style guides, and edge cases.
Review other raters' work, calibrate standards, and develop or refine grading guidelines.
Assess search relevance and provide feedback in live sessions.
At least 5 years of professional experience in rating, review, editing, fact-checking, trust and safety, or a relevant professional domain.
Native or near-native written English proficiency.
Experience grading or reviewing expert-level answers against rubrics, policies, or style guides.
Experience leading or reviewing other raters, such as as a QA lead, rubric owner, or reviewer of reviewers.
An advanced degree, professional license, or deep working expertise in the domain you grade.
Hands-on AI data annotation experience, search-quality rating, or experience writing and maintaining grader guidelines is a plus.
Compensation is $50 to $120 per hour, depending on the project, field, and expertise level. Payment is based on hours worked.
Fully remote. Projects typically run from one week to several months.

Huntress
USA
What We Do: Cybercrime is growing, and more businesses are getting hit by threats that used to target only the biggest organizations. That pushes defenders like us to operate at the highest level, and it deepens our need for good people who want to make a meaningful impact. Founded in 2015 by former NSA cyber…

Sayari
USA
About Sayari: Sayari is the judgment infrastructure for trustworthy AI in economic security and commercial risk. The Sayari Commercial World Model resolves 11.7B+ primary-source records from 250+ jurisdictions forming the ground truth of global commerce. A Judgment Ontology, encoding over a decade of investigative tradecraft, and Superconductor, an agentic orchestration platform, deliver AI that reasons…