AI Quality Analyst – Portuguese (Remote)

About the role

As an AI Quality Analyst you will evaluate a new personalization feature for Gemini, assessing how the model leverages data from past Gemini conversations, Gmail, Google Search, and YouTube to generate more relevant and helpful responses. The position blends creative prompt design with rigorous analysis of model outputs, requiring both imagination and analytical precision.

Key responsibilities

• Design and execute multi‑turn prompts based on your personal experiences to test model capabilities.
• Review side‑by‑side model responses and rank them on grounding, integration, and helpfulness.
• Write clear, concise rationales referencing specific turn numbers.
• Provide detailed annotations and constructive feedback on personalization quality.
• Collaborate with a global 24‑hour operations team while working remotely.

Required profile

• High proficiency in reading and writing Portuguese.
• Willingness to use your primary personal Google account and enable relevant data sources.
• Full‑time availability in your local time zone.
• Strong analytical thinking for evaluating nuanced AI responses.
• Self‑motivated, able to work independently in a remote setting.
• Excellent written communication skills.

Required skills

• Prompt engineering for AI evaluation.
• Ability to assess personalization quality and identify incorrect inferences.
• Meticulous attention to detail for spotting subtle differences in model outputs.
• Reliable desktop/laptop setup with a good internet connection.