Remote Contract: Video QA & Rubric Evaluator

AuraOne; United States; On-site; USD 207,000 - 276,000; Full time

Benefits offered by this job

Health insurance

Job summary

AuraOne is seeking an independent contractor to evaluate UK-based video producers and editors outputs according to a versioned rubric. You will assign severity tags, compare paired responses, and provide structured, reasoned feedback for retraining the model.

Responsibilities include labeling edge cases, flagging hallucinations or unsafe content, and documenting recurring failure modes for ongoing improvement. Remote work is allowed with flexible scheduling.

Qualifications

• Experience evaluating or annotating video producers editors content for UK-based work.
• Ability apply multi-page rubrics consistently across long batches.
• Ability to write clear reasoning naming the issue and rubric clause.
• Strong attention to detail and identify when prompt is the problem.
• Reliable async availability for at least 10 hours per week.

Responsibilities

• Evaluate UK-based video producers editors evaluation model outputs against a versioned rubric and assign severity tags.
• Compare paired responses and select the stronger answer with a written rationale.
• Label hallucinations, instruction-following failures, and unsafe content with structured tags.
• Capture ambiguous prompts and route to program team for rubric updates.
• Maintain reviewer-quality scores by calibrating against gold-standard examples weekly.
• Document recurring failure modes for model training improvements.

Skills

Evaluation experience

Rubric consistency

Clear reasoning

Attention to detail

Async availability

Job description

AuraOne is seeking an independent contractor to evaluate UK-based video producers and editors outputs according to a versioned rubric. You will assign severity tags, compare paired responses, and provide structured, reasoned feedback for retraining the model.

Responsibilities include labeling edge cases, flagging hallucinations or unsafe content, and documenting recurring failure modes for ongoing improvement. Remote work is allowed with flexible scheduling.