Remote Voice Agent Evaluation Specialist

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

Job summary

AuraOne is seeking a remote Voice Agent Conversation Evaluator contractor to review and rate AI-generated prompts and responses against our quality rubric. You will compare paired outputs, label edge cases, and provide written rationales to help retrain the model.

Responsibilities include tagging hallucinations, instruction-following failures, and unsafe content with structured categories, calibrating against gold-standard examples, and documenting recurring issues for improvement.

Qualifications

• Prior evaluation, annotation, or human-rater experience on voice agent 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 ability to flag when a prompt is the problem.
• Reliable async availability for at least 10 hours per week.

Responsibilities

• Evaluate voice agent conversation evaluation model outputs against a versioned rubric and assign severity tags for Voice Agent Conversation 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 output evaluation

Rubric-based annotation

Severity tagging

Inter-rater calibration

Voice Agent Conversation evaluation

Speech evaluation

Voice QA

Audio review

Voice

Agent

Job description

AuraOne is seeking a remote Voice Agent Conversation Evaluator contractor to review and rate AI-generated prompts and responses against our quality rubric. You will compare paired outputs, label edge cases, and provide written rationales to help retrain the model.

Responsibilities include tagging hallucinations, instruction-following failures, and unsafe content with structured categories, calibrating against gold-standard examples, and documenting recurring issues for improvement.