Web Browsing Evaluator

AI Trainer Jobs; United States; Remote; USD 85,962,000 - 171,924,000; Full time

Job summary

AuraOne is seeking a Web Browsing Evaluator to remotely review prompts and responses using a structured rubric. You will compare paired outputs, label edge cases, and provide actionable feedback for retraining the model.

Responsibilities include tagging issues, calibrating scores, and ensuring consistency across long task batches with reliable async availability.

Qualifications

• Experience evaluating prompts and model outputs against rubrics.
• Ability to annotate with structured feedback and rationale.
• Familiarity with inter-rater calibration and drift reporting.

Responsibilities

• Evaluate web browsing model outputs against a versioned rubric.
• Assign severity tags for Web Browsing Evaluator tasks.
• Compare paired responses and provide written rationales.
• Label hallucinations and safety issues with structured tags.
• Flag ambiguous prompts and suggest rubric updates.

Skills

Attention to detail

Written reasoning

Async availability

Inter-rater calibration

Content moderation experience

Web Browsing evaluation

Rubric-based annotation

Job description

Web Browsing Evaluator is a remote evaluation track for reviewing web browsing evaluation prompts and responses against AuraOne's quality rubric. Reviewers compare paired outputs, label edge cases, and write the kind of structured feedback the modeling team can use to retrain.

Category: Search, Web & Browser Agents · Pay: $30–$60 / hr · Location: Remote — US-eligible · Contractor

Web Browsing Evaluator is a remote evaluation track for reviewing web browsing evaluation prompts and responses against AuraOne's quality rubric.

About the role

Web Browsing Evaluator is a remote evaluation track for reviewing web browsing evaluation prompts and responses against AuraOne's quality rubric. Reviewers compare paired outputs, label edge cases, and write the kind of structured feedback the modeling team can use to retrain.
AI data reviewers help turn web browsing evaluation outputs into auditable labels, rationales, and regression cases for AuraOne Human Data.
Review browser automation and multi-step search agents. Check whether sources hold up.

Responsibilities

• Evaluate web browsing evaluation model outputs against a versioned rubric and assign severity tags for Web Browsing 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.

What you should bring

• Prior evaluation, annotation, or human-rater experience on web browsing evaluation or adjacent content for Web Browsing Evaluator work.
• 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.

Example tasks

• Compare two web browsing evaluation model responses to the same prompt and pick the stronger one with rationale.
• Tag an unsafe response with the correct policy category and severity.
• Audit a 50-row batch for rubric consistency and report drift to the program lead.
• Propose a rubric clarification after spotting a recurring failure mode.

Useful experience

• Background in linguistics, content moderation, or trust & safety review.
• Experience with inter-rater agreement metrics and calibration cycles.
• Domain expertise that lets you spot subject-matter errors automated checks miss.

Compensation and schedule

Hourly rate confirmed after the interview process.
Expected arrangement: contractor , with program-defined task volume and review pacing. Placement depends on current program demand and reviewer confirmation.

Skills used in matching

• Model output evaluation
• Rubric-based annotation
• Severity tagging
• Inter-rater calibration
• Web Browsing evaluation
• AI model evaluation