AI Trainer, Response Evaluation

HireArt is helping An AI Firm find AI Trainers to evaluate model responses. In this role you’ll contribute to the advancement of AI technology by helping to refine and improve AI algorithms.

Responsibilities:

• Evaluate two model responses per prompt based on predetermined criteria by answering several questions.
• Use provided criteria provided to determine whether model response are harmless, honest, and helpful.
• Compare the model responses to see which one is better overall and by how much.
• Maintain a high level of accuracy and consistency in evaluations.

Role Details:

• It typically takes 5 minutes to evaluate each set of prompt and model responses.
• There is no limit to how many responses you can evaluate
• Compensation is based on the hours spent actively working in the evaluation tool.

Requirements:

• Native fluency in English
• Reside in the United States
• At least 18 years old
• Have access to a laptop or PC and stable internet connection
• Previous experience with data annotation, AI training/AI annotation, content review, search quality rating, or similar

Ideal Qualifications:

• At least 6 months of experience with data annotation, AI training/AI annotation, content review, search quality rating, or similar
• Able to start immediately
• Able to work 25+ hours per week
• Previous experience in AI annotation or training
• Previous experience doing evaluation or grading work, such as a teacher, professor, teacher’s assistant, content reviewer, data labeler, search rank evaluator, or editor

Commitment: This is a 6-week, independent contractor position staffed via HireArt. This role will be remote and open to candidates who reside and are legally authorized to work in the United States.

HireArt values diversity and is an Equal Opportunity Employer. We are interested in every qualified candidate who is eligible to work in the United States. Unfortunately, we are not able to sponsor visas.