Remote Evaluator: Preference Taxonomy & Model Quality

AuraOne; United States; On-site; USD 34,440 - 55,104; Part time

Job summary

AuraOne is seeking a remote Preference Taxonomy Reward Model Evaluator contractor to review model outputs against a versioned rubric, label edge cases, and provide structured feedback for retraining. You will compare paired responses, tag issues, and route ambiguous prompts for rubric updates.

Ideal candidates have prior evaluation or annotation experience, strong attention to detail, and the ability to work asynchronously for at least 10 hours per week.

Qualifications

• Prior evaluation, annotation, or human-rater experience on preference taxonomy reward model evaluation or adjacent content.
• Clear written reasoning that names the issue and the rubric clause being applied.
• Strong attention to detail and ability to flag prompt issues.

Responsibilities

• Evaluate outputs against a versioned rubric and assign severity tags.
• Compare paired responses and select the stronger answer with rationale.
• Label hallucinations, instruction-following failures, and unsafe content with structured tags.
• Capture ambiguous prompts and route them for rubric updates.
• Calibrate reviewer-quality scores against gold-standard examples weekly.
• Document recurring failure modes to inform future training.

Skills

Model evaluation

Rubric annotation

Severity tagging

Inter-rater calibration

RLHF

Rater calibration

Job description

AuraOne is seeking a remote Preference Taxonomy Reward Model Evaluator contractor to review model outputs against a versioned rubric, label edge cases, and provide structured feedback for retraining. You will compare paired responses, tag issues, and route ambiguous prompts for rubric updates.

Ideal candidates have prior evaluation or annotation experience, strong attention to detail, and the ability to work asynchronously for at least 10 hours per week.