Remote Robot Incident Evaluation Specialist

AuraOne; United States; On-site; USD 34,440 - 55,104; Full time

Job summary

AuraOne is seeking a Remote Robot Incident Replay Task Evaluator to review model outputs against our quality rubric and generate auditable labels, rationales, and regression cases for retraining.

As an independent contractor, you will compare paired responses, assign severity tags, label edge cases, and document recurring failure modes to guide future training. US-eligible work and ~10 hours weekly are expected, with compensation discussed after interview.

Qualifications

• Prior evaluation, annotation, or human-rater experience on robot incident replay task evaluation or adjacent content for Robot Incident Replay Task Evaluator work.
• 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 task evaluation model outputs against a versioned rubric and assign severity tags for Robot Incident Replay Task 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.
• Document recurring failure modes so the modeling team can target them in the next training run.

Skills

Model evaluation

Rubric annotation

Severity tagging

Inter-rater calibration

Robot evaluation

Safety review

Job description

AuraOne is seeking a Remote Robot Incident Replay Task Evaluator to review model outputs against our quality rubric and generate auditable labels, rationales, and regression cases for retraining.

As an independent contractor, you will compare paired responses, assign severity tags, label edge cases, and document recurring failure modes to guide future training. US-eligible work and ~10 hours weekly are expected, with compensation discussed after interview.