Remote Web Browsing Evaluator - QA & Feedback

AuraOne Human Data; United States; Remote; USD 21,000 - 34,000; Full time

Job summary

AuraOne Human Data is seeking a Web Browsing Evaluator for a remote role. You will review prompts and responses against rubrics, label edge cases, and provide structured feedback to retrain models.

Responsibilities include evaluating model outputs, comparing paired responses, labeling hallucinations and unsafe content, and surfacing prompts for rubric updates. Requires reliable async availability for at least 10 hours per week.

Qualifications

• Experience in evaluation, annotation, or human-rater tasks for web browsing.
• Ability to consistently apply rubrics across long batches.
• Clear written reasoning that cites rubric clauses.
• Strong attention to detail and the ability to flag problematic prompts.
• Reliable async availability for at least 10 hours per week.

Responsibilities

• Evaluate web browsing model outputs against a versioned rubric and assign severity tags.
• Compare paired responses and select the stronger answer with a rationale.
• Label hallucinations, instruction-following failures, and unsafe content with structured tags.
• Capture ambiguous prompts and route to the program team for rubric updates.
• Maintain reviewer-quality scores by calibrating against gold-standard examples.
• Document recurring failures to guide next training runs.

Skills

Evaluation experience

Annotation

Attention to detail

Written reasoning

Job description

AuraOne Human Data is seeking a Web Browsing Evaluator for a remote role. You will review prompts and responses against rubrics, label edge cases, and provide structured feedback to retrain models.

Responsibilities include evaluating model outputs, comparing paired responses, labeling hallucinations and unsafe content, and surfacing prompts for rubric updates. Requires reliable async availability for at least 10 hours per week.