Knowledge Retrieval QA Reviewer & Rubric Architect

AuraOne, Inc.; United States; Remote; USD 34,000 - 55,000; Full time

Benefits offered by this job

Remote work (US-eligible)

Contractor employment

Job summary

AuraOne, Inc. is seeking a Knowledge Retrieval Research QA Specialist as a remote contractor to review prompts and model outputs against AuraOne's quality rubric. You will label edge cases, write structured feedback, and help retrain the evaluation system.

Responsibilities include comparing paired responses, labeling hallucinations and safety issues, and calibrating scores against gold standards. This role requires strong written reasoning and reliable async availability for at least 10 hours

Qualifications

• Experience in evaluation, annotation, or human-rater work on knowledge retrieval QA or related content.
• Ability to apply multi-page rubrics consistently across long batches.
• Clear written reasoning that names the issue and cites the rubric clause.
• Strong attention to detail and ability to flag when prompts are the problem.
• Reliable async availability for at least 10 hours per week.

Responsibilities

• Evaluate outputs against rubric and assign severity tags.
• Compare paired responses and select the stronger one with rationale.
• Label hallucinations, instruction-following failures, and unsafe content with structured tags.
• Capture ambiguous prompts and route to program team for rubric updates.
• Calibrate reviewer scores against gold-standard examples weekly.
• Document recurring failure modes for model improvement.

Skills

Model output evaluation

Rubric-based annotation

Severity tagging

Inter-rater calibration

Knowledge Retrieval Research QA eval

Web research

Source grounding

Browser automation

Knowledge

Retrieval

Job description

AuraOne, Inc. is seeking a Knowledge Retrieval Research QA Specialist as a remote contractor to review prompts and model outputs against AuraOne's quality rubric. You will label edge cases, write structured feedback, and help retrain the evaluation system.

Responsibilities include comparing paired responses, labeling hallucinations and safety issues, and calibrating scores against gold standards. This role requires strong written reasoning and reliable async availability for at least 10 hours