Remote Generalist Evaluation Expert (UK/Europe)

AuraOne; United States; Remote; USD 69,000 - 96,000; Part time

Benefits offered by this job

Remote work

Contractor arrangement

Job summary

AuraOne seeks a Generalist Expert (UK/Europe) to remotely review prompts and responses against its quality rubric. You will compare outputs, label issues, and provide structured feedback for retraining efforts.

As a contractor, you will evaluate model results, identify edge cases, and write clear rationales that map to rubric clauses. Reliable async availability of at least 10 hours per week is expected.

Qualifications

• Prior evaluation, annotation, or human-rater experience on generalist evaluation or adjacent content for Generalist Expert (UK/Europe) 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.

Responsibilities

• Evaluate generalist evaluation model outputs against a versioned rubric and assign severity tags for Generalist Expert (UK/Europe) 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.

Skills

Model output evaluation

Rubric-based annotation

Severity tagging

Inter-rater calibration

Generalist evaluation

Job description

AuraOne seeks a Generalist Expert (UK/Europe) to remotely review prompts and responses against its quality rubric. You will compare outputs, label issues, and provide structured feedback for retraining efforts.

As a contractor, you will evaluate model results, identify edge cases, and write clear rationales that map to rubric clauses. Reliable async availability of at least 10 hours per week is expected.