Remote AI Generalization Review Specialist

AuraOne, Inc.; United States; Remote; USD 55,000 - 83,000; Full time

Job summary

AuraOne, Inc. is seeking a Capability Generalization Failure Analysis Reviewer for remote evaluation. You will assess prompts and model outputs against AuraOne's quality rubric, compare paired responses, and label edge cases with structured feedback to train the team.

Your work involves annotating failures, naming issues with rubric references, and maintaining calibration across batches for ongoing quality improvements.

Qualifications

• Prior evaluation, annotation, or human-rater experience on capability generalization failure analysis 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 capability generalization failure analysis evaluation model outputs against a versioned rubric and assign severity tags.
• Compare paired responses and pick the stronger answer with 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

Capability generalization

Job description

AuraOne, Inc. is seeking a Capability Generalization Failure Analysis Reviewer for remote evaluation. You will assess prompts and model outputs against AuraOne's quality rubric, compare paired responses, and label edge cases with structured feedback to train the team.

Your work involves annotating failures, naming issues with rubric references, and maintaining calibration across batches for ongoing quality improvements.