Remote Pairwise Preference Evaluator & Feedback Specialist

AuraOne; United States; On-site; USD 27,552 - 55,104; Part time

Job summary

AuraOne is seeking a remote, contractor-based Pairwise Preference Reward Model Evaluator to review prompts and responses against the company's quality rubric. You will compare paired outputs, label edge cases, and draft structured feedback to help retrain the modeling team.

The role emphasizes careful reasoning, consistency across long review batches, and clear documentation of errors. You will work asynchronously for at least 10 hours weekly, with compensation discussed after the interview.

Qualifications

• Prior evaluation, annotation, or human-rater experience on pairwise reward model evaluation or adjacent content.
• 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 pairwise preference reward model evaluation model outputs against a versioned rubric and assign severity tags for Pairwise Preference Reward Model 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

Model output evaluation

Rubric-based annotation

Severity tagging

Inter-rater calibration

Pairwise evaluation

RLHF

Rater calibration

Job description

AuraOne is seeking a remote, contractor-based Pairwise Preference Reward Model Evaluator to review prompts and responses against the company's quality rubric. You will compare paired outputs, label edge cases, and draft structured feedback to help retrain the modeling team.

The role emphasizes careful reasoning, consistency across long review batches, and clear documentation of errors. You will work asynchronously for at least 10 hours weekly, with compensation discussed after the interview.