The Work
You will create and assess specialized AI training tasks across market research and analytics, media production and direction, and editorial and technical communications. Your work will combine subject-matter judgment with structured review and repeated feedback.
You will develop realistic workplace scenarios, solution sets that represent strong answers, and rubrics used to score AI outputs. Reviews will focus on accuracy, terminology, quality, logical consistency, formatting, and adherence to the rubric.
- Design difficult domain tasks for your area of professional experience.
- Create realistic workplace scenarios, solution sets, and evaluation standards.
- Score AI-generated outputs consistently against accuracy, terminology, and quality criteria.
- Refine work through feedback and repeated critique cycles.
- At senior levels, review peer outputs, resolve unclear edge cases, and identify rubric discrepancies.
What It Pays And Takes
This is a contractor, part-time engagement. The listing does not provide a pay rate. The role is open worldwide and requires English fluency.
- Engagement length: 12 weeks.
- Availability: at least 4 hours per day and up to 40 hours per week.
- Structured time requirement: 20+ hours per week.
- Schedule requirement: 4 hours of overlap with Pacific Standard Time.
- Experience level: entry level.
- Professional background: 3 to 15+ years of demonstrable experience in market research and analytics, media production and direction, or editorial and technical communications.
- Required skills: strong written communication, documentation, objective critique, and close attention to detail.
- Education: a bachelor's degree or equivalent practical experience in any field.
- Helpful but not required: AI evaluation, data annotation, content review, quality assurance, or related analytical experience.
How It Works
Apply on OpenTrain with your resume and then complete the application on the hiring site.
About AI Training Work
AI training is the human work behind many AI systems, including writing tasks, rating model responses, and checking outputs against clear standards. People with professional experience help make evaluations accurate by applying informed judgment to specialized subjects.