Remote Browser Automation & Tool-Use Evaluator

AuraOne; United States; On-site; USD 34,000 - 83,000; Part time

Job summary

AuraOne is seeking a Browser Automation and Tool-Use Specialist for a remote evaluation track. Review prompts and responses against AuraOne's quality rubric, label edge cases, and provide structured feedback for retraining.

As a CONTRACTOR, you will compare paired outputs, apply severity tags, and document recurring failure modes to help the modeling team improve future iterations. Remote work is available to US-eligible contractors.

Qualifications

• Experience in evaluation, annotation, or human-rater work on browser automation and tool-use evaluation or adjacent content.
• Ability to apply 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 model outputs against a versioned rubric and assign severity tags.
• Compare paired responses and write a 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 weekly.
• 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

Browser Automation and Tool Use eval

Software engineering and computer use

AI evaluation

Rubric writing

Expert review

Job description

AuraOne is seeking a Browser Automation and Tool-Use Specialist for a remote evaluation track. Review prompts and responses against AuraOne's quality rubric, label edge cases, and provide structured feedback for retraining.

As a CONTRACTOR, you will compare paired outputs, apply severity tags, and document recurring failure modes to help the modeling team improve future iterations. Remote work is available to US-eligible contractors.