AI Trainer – Portuguese (Remote, $15/hr)

About the role

We are looking for a Portuguese‑speaking AI Trainer to evaluate and improve a personalization feature for a large language model. This short‑term, part‑time contract involves testing how the model uses personal context to generate relevant, accurate responses.

Key responsibilities

• Design and execute multi‑turn conversational prompts that require the model to leverage personal context.
• Assess the model’s ability to incorporate past conversation data and user activity into its answers.
• Evaluate responses for intent alignment, grounding issues, flawed inferences, or hallucinations.
• Perform side‑by‑side comparisons and rank model outputs based on usefulness, clarity, and overall experience.
• Write clear rationales for rankings, referencing specific conversation turns.
• Extract and verify debug information to confirm correct use of summaries and data sources.
• Maintain data hygiene by deleting evaluation conversations after completion.

Required profile

• Experience in data annotation, AI quality evaluation, content moderation, or related fields.
• Fluent reading and writing skills in Portuguese.
• Willingness to use a personal Google account and enable personal data sources for assessment.
• Strong analytical thinking, attention to detail, and ability to work independently in a remote setup.

Required skills

• Prompt engineering
• Data annotation
• AI quality evaluation

What we offer

• Hourly compensation ranging from $11 to $15.
• Flexible part‑time schedule (20–40 hrs/week) over a two‑month period.
• Remote work with a stable internet connection requirement.