New AI Voice Evaluation Specialist (Fully Remote)

Innodata; United States; Remote; USD 28,000 - 37,000; Full time

Job summary

Innodata in the United States is seeking a detail-focused evaluator to review assigned conversational scenarios and roleplay with two AI models.

You will record voice interactions, evaluate responses across five quality dimensions, and provide clear, objective rationale with specific examples from both conversations. The position requires a Bachelor's degree and strong listening and writing skills, with the ability to follow guidelines.

Qualifications

• Bachelor's degree required with strong attention to detail and writing skills.
• Ability to follow guidelines and speak naturally in roleplay scenarios.

Responsibilities

• Review assigned scenario and roleplay natural conversations with two AI models.
• Record voice interactions and evaluate responses across five quality dimensions.
• Identify and categorize error clusters, compare model performance, and select the stronger model.
• Write clear, objective rationale with specific examples from both conversations.

Skills

Attention to detail

Listening

Writing

Education

Bachelor's degree

Job description

• Review assigned scenario and roleplay natural conversations with two AI models, ensuring consistency in approach and turn count.
• Record voice interactions and evaluate responses across five quality dimensions.

Evaluation:

• Identify and categorize error clusters, compare model performance, and select the stronger model.
• Write clear, objective rationale with specific examples from both conversations.

Requirements:

• Bachelor's degree and strong attention to detail, listening, and writing skills.
• Ability to follow guidelines and speak naturally in roleplay scenarios.

Compensation:

• Hourly rate of $20–$27, based on experience, skills, and qualifications.

Innodata

Innodata is a global data engineering company that enables responsible AI advancement by providing data, evaluation frameworks, and human expertise. With a 36+ year legacy, they deliver high-quality data and outcomes for AI builders and adopters.