Remote AI Data Evaluator & Rubric Reviewer

AuraOne; United States; On-site; USD 34,440 - 61,992; Full time

Job summary

AuraOne is seeking a remote Contractor to join its AI data review team as a Customer Support Agent Evaluator, focusing on evaluating model outputs against the quality rubric. You will compare paired responses, label edge cases, and write structured feedback the modeling team can use to retrain.

Reviewers also calibrate scores against gold standards and document recurring failure modes to improve future training runs.

Qualifications

• Prior evaluation, annotation, or human-rater experience on customer support agent evaluation or adjacent content for Customer Support Agent Evaluator work.
• Comfort applying multi-page rubrics consistently across long batches.
• Clear written reasoning that names the issue and the rubric clause being applied.
• Strong attention to detail and the ability to flag when a prompt itself is the problem.
• Reliable async availability for at least 10 hours per week.

Responsibilities

• Evaluate customer support agent evaluation model outputs against a versioned rubric and assign severity tags for Customer Support Agent Evaluator assignments.
• Compare paired responses and pick the stronger answer with a written rationale.
• Label hallucinations, instruction-following failures, and unsafe content with structured tags.
• Capture ambiguous prompts and route them back to the program team for rubric updates.
• Maintain reviewer-quality scores by calibrating against gold-standard examples each week.
• Document recurring failure modes so the modeling team can target them in the next training run.

Job description

AuraOne is seeking a remote Contractor to join its AI data review team as a Customer Support Agent Evaluator, focusing on evaluating model outputs against the quality rubric. You will compare paired responses, label edge cases, and write structured feedback the modeling team can use to retrain.

Reviewers also calibrate scores against gold standards and document recurring failure modes to improve future training runs.