Remote Fraud Reasoning Evaluator & Rubric Annotator

AuraOne; United States; On-site; USD 3,444,000 - 5,510,400; Full time

Job summary

AuraOne is seeking a Fraud Detection Reasoning Evaluator for a remote contractor role to review prompts and model outputs, apply the quality rubric, and produce auditable labels, rationales, and regression cases for the human data team.

You will compare paired responses, assign severity tags, label hallucinations and policy breaches, flag ambiguous prompts, and help calibrate scores weekly while ensuring US-eligibility and timely async availability.

Qualifications

• Prior evaluation, annotation, or human-rater experience on fraud detection reasoning evaluation or adjacent content for Fraud Detection Reasoning 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 fraud detection reasoning evaluation model outputs against a versioned rubric and assign severity tags for Fraud Detection Reasoning 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.

Skills

Rubric annotation

Severity tagging

Inter-rater calibration

Fraud evaluation

Model evaluation

Job description

AuraOne is seeking a Fraud Detection Reasoning Evaluator for a remote contractor role to review prompts and model outputs, apply the quality rubric, and produce auditable labels, rationales, and regression cases for the human data team.

You will compare paired responses, assign severity tags, label hallucinations and policy breaches, flag ambiguous prompts, and help calibrate scores weekly while ensuring US-eligibility and timely async availability.