Portuguese AI Personalization Quality Analyst

About OpenTrain

OpenTrain is the #1 platform for finding and building careers in AI training and data labeling. We help freelancers discover projects, build a single portfolio of AI-training work, and grow durable freelance careers doing the human work that trains modern AI.

OpenTrain AI is the hiring and contracting organization for this role. Creating an OpenTrain account is free and central to applying and tracking your work.

About AI Training Work

AI training (data labeling / annotation / human feedback) is the human side of building AI: people create, review, and rate examples that teach models how to respond. This role focuses on personalization evaluation — shaping how generative models use personal context and conversation history.

This kind of work is often 100% remote, flexible, and accessible without prior technical experience; strong language skills and careful judgment are the most important qualifications here.

The Role

You will evaluate a personalization feature for a generative model using your personal account activity and conversation history. The work emphasizes designing short multi-turn prompts from your own experiences, assessing grounding and helpfulness, comparing side-by-side model outputs, and producing clear written rationales.

This is a contractor, part-time role requiring 20+ hours per week. Pay is USD $15 per hour. Work is remote and open worldwide; Portuguese fluency is required.

• Job type: Contractor, Part-time
• Time requirement: 20+ hours/week
• Pay: $15 USD per hour
• Language: Portuguese reading and writing fluency required
• Data type: Text — evaluation ratings and written rationales

What You'll Do

Your day-to-day includes creating prompts drawn from your own personal context, judging how naturally the model integrates personal information, and documenting where models make incorrect inferences or hallucinate facts.

• Design 1–5 turn prompts based on your personal experiences and account activity.
• Evaluate model responses for grounding, naturalness, helpfulness, and overnarration.
• Compare two model responses side-by-side and choose the more useful answer.
• Write structured, precise rationales citing specific conversation turns and errors.
• Maintain data hygiene by deleting evaluation conversations after use.

Requirements

Candidates must meet the stated language and evaluation skills and be able to work remotely with a reliable computer and internet connection.

• Fluent Portuguese reading and writing ability (required).
• Strong analytical judgment for nuanced AI response evaluation.
• Experience designing creative multi-turn prompts or testing conversational systems.
• Detail-oriented: able to produce clear, structured rationales and spot subtle grounding issues.
• Familiarity with data annotation, AI quality evaluation, or content moderation is preferred.
• Ability to delete evaluation conversations and follow data-hygiene instructions.

Preferred Background

We welcome applicants with analytical training or experience that helps evaluate nuanced language and behavior. These backgrounds are helpful but not required.

• Policy, law, ethics, or journalism experience — useful for careful judgment.
• Linguistics or language-related degrees/experience (helpful).
• Comfortable working local hours with some overlap for a global operations team.