Remote Robot Incident Replay Failure Case Evaluator

AI Trainer Jobs; United States; Remote; USD 34,000 - 55,000; Part time

Job summary

AuraOne is seeking a remote contractor to review robot incident replay failure case evaluation prompts and responses. You will compare outputs, label edge cases, and provide structured feedback to retrain the model with a versioned rubric.

Responsibilities include evaluating model outputs, tagging issues, and routing ambiguous prompts back to the team for rubric updates. Strong written reasoning and reliability for at least 10 hours per week are essential.

Qualifications

• Prior evaluation, annotation, or human-rater experience on robot incident replay evaluation or adjacent content.
• 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 robot incident replay failure case evaluation model outputs against a rubric and assign severity tags.
• Compare paired responses and write a justification for selecting the stronger answer.
• Label hallucinations, instruction-following failures, and unsafe content with structured tags.
• Capture ambiguous prompts and route to program team for rubric updates.

Skills

Evaluation experience

Inter-rater calibration

Linguistics background

Attention to detail

Job description

AuraOne is seeking a remote contractor to review robot incident replay failure case evaluation prompts and responses. You will compare outputs, label edge cases, and provide structured feedback to retrain the model with a versioned rubric.

Responsibilities include evaluating model outputs, tagging issues, and routing ambiguous prompts back to the team for rubric updates. Strong written reasoning and reliability for at least 10 hours per week are essential.