Remote Japanese Bilingual Writer & Evaluator

AI Trainer Jobs; United States; Remote; USD 21,000 - 29,000; Part time

Benefits offered by this job

Remote contract

Flexible hours

Job summary

AI Trainer Jobs is seeking a remote contractor for Bilingual Writer - Japanese (Japan) to review Japanese generalist evaluation prompts and responses against AuraOne's quality rubric. You will compare paired outputs, label edge cases, and provide structured feedback to retrain models.

The role emphasizes remote, async work with weekly calibration against gold-standard examples. Compensation is $18 per hour, with flexible task volume and program-defined pacing.

Qualifications

• Prior evaluation, annotation, or human-rater experience on japanese generalist evaluation or adjacent content.
• Clear written reasoning that names the issue and the rubric clause being applied.
• Reliable async availability for at least 10 hours per week.
• Strong attention to detail and ability to flag when a prompt is the problem.

Responsibilities

• Evaluate japanese generalist evaluation model outputs against a versioned rubric and assign severity tags.
• Compare paired responses and select the stronger one with a written rationale.
• Label hallucinations, instruction-following failures, and unsafe content with structured tags.
• Capture ambiguous prompts and escalate to the program team for rubric updates.
• Maintain reviewer-quality scores by calibration against gold-standard examples.

Skills

Japanese evaluation experience

Annotation & labeling

Inter-rater calibration

Rubric application

Job description

AI Trainer Jobs is seeking a remote contractor for Bilingual Writer - Japanese (Japan) to review Japanese generalist evaluation prompts and responses against AuraOne's quality rubric. You will compare paired outputs, label edge cases, and provide structured feedback to retrain models.

The role emphasizes remote, async work with weekly calibration against gold-standard examples. Compensation is $18 per hour, with flexible task volume and program-defined pacing.