AI trainer

What you would do

• Review and provide detailed feedback on AI-generated examples and outputs
• Write and refine prompts and tasks designed to test AI reasoning capabilities
• Annotate and collaborate on dataset improvements with other domain experts
• Document findings and report progress to project teams
• Assess AI performance in realistic scenarios across your field of expertise

Who they want

• Deep expertise in a specialized domain: science, law, engineering, writing, education, or similar
• Exceptional written communication and attention to detail
• Proven critical thinking and ability to articulate nuanced feedback
• Background with labeling, instructional content, scholarly work, or domain-specific writing (optional)
• Comfort working independently and remotely with self-directed project adjustment

Main skills

What the interview asks about

1. 1.Domain knowledge application

   Your job is to spot when AI outputs miss domain-specific nuance that matters in real applications, so the interviewer checks if you understand your field deeply enough.

   For example: “A financial AI model generates advice on loan structuring that omits tax implications. How would you identify this gap, and what prompt or feedback would you provide to address it?”

2. 2.Prompt design for AI reasoning

   You shape what AI attempts by writing tasks and questions, so interviewers want to see if you can craft prompts that challenge systems in useful ways.

   For example: “You notice the AI excels at summarization but struggles with multi-step legal analysis. What refinements would you make to your prompts to push it toward deeper reasoning?”

3. 3.Quality standards and documentation

   You're responsible for ensuring deliverables meet project standards and communicating findings clearly, which requires structured thinking and clear writing.

   For example: “You find that 30% of the AI outputs on a dataset fail to meet accuracy standards. How would you document the patterns you observe and communicate them to your stakeholders?”

4. 4.Independent work in remote settings

   Without in-person feedback or daily oversight, you must manage your own workflow, interpret guidelines, and adapt when context shifts.

   For example: “Midway through a project, guidelines change to emphasize a different quality metric. Walk through how you'd adjust your annotation and feedback approach.”

5. 5.Collaborative annotation and feedback

   You work alongside other experts to improve datasets, so the interviewer checks if you can accept feedback, align on standards, and contribute constructively.

   For example: “Another annotator marks an output as acceptable, but you think it contains a subtle error. How would you discuss and resolve the disagreement?”

A task you may get

Review three AI model outputs in your domain and provide: a brief quality assessment for each, specific feedback on errors or gaps, and a revised prompt designed to prevent each identified issue.

How to prepare

• Study recent AI model failures or limitations in your domain, noting what types of errors occur and what context the model lacks
• Practice articulating domain-specific feedback clearly and concisely, focusing on actionable improvements rather than vague critiques
• Reflect on a time you evaluated someone else's work: how did you balance honesty with constructiveness, and how would you handle disagreement?

The facts

Pay

$100–180/hr

Open to

Bangladesh, Hong Kong, India, Indonesia, Japan, Kazakhstan, Kyrgyzstan, Malaysia, Pakistan, Philippines, Singapore, Sri Lanka, Taiwan, Thailand, Uzbekistan, Vietnam, Austria, Belarus, Belgium, Denmark, France, Germany, Greece, Italy, Netherlands, Portugal, Russia, Spain, Switzerland, United Kingdom, Argentina, Brazil, Chile, Colombia, Mexico, Peru, Algeria, Bahrain, Egypt, Iraq, Jordan, Kuwait, Lebanon, Libya, Morocco, Oman, Palestine, Qatar, Saudi Arabia, Tunisia, United Arab Emirates, United States, Canada, Nigeria, Kenya, South Africa, Ghana, Ethiopia

Field

Software Engineering

Role type

Generalist

Posted

7/30/2026

Places left

10

We wrote this page from the public micro1 listing. It may be out of date, so read the full posting before you apply.