AI Quality Analyst (Gemini) - Chinese

• Evaluate a new personalization feature for Gemini, assessing how well the model uses information from past conversations, Gmail, Google Search, and YouTube activity to make responses more relevant and helpful.
• Design prompts from personal experiences.
• Assess the quality of the model's personalized responses, evaluating dimensions like Grounding, Integration, and Helpfulness.
• Design and execute multi-turn conversational prompts (typically 1-5 turns) that require the AI to utilize personal information and experiences.
• Analyze responses for Grounding issues, ensuring claims about you are supported by evidence and not flawed inferences or hallucinations.
• Rigorously evaluate and stack-rank two model responses side-by-side (SxS) to determine which is overall more helpful, easy to use, and enjoyable.
• Write clear, defensible rationales for comparisons, explicitly referencing where issues or positive aspects occurred in the conversation.
• Extract and verify "Debug Info" from the model to confirm that chat summaries and data sources were properly utilized.
• Maintain strict data hygiene by deleting evaluation conversations to prevent them from polluting future chat history.

• Ability to read and write in Chinese with a high degree of competence.
• Experience in designing creative, multi-turn starting prompts based on personal context to thoroughly test the model's capabilities.
• Understanding of personalization concepts, including the ability to identify incorrect personalization, poor inferences, and forced connections.
• Ability to review Side-by-Side (SxS) model responses and spot subtle differences in naturalness and overnarrating.
• Superior ability to write clear, concise, and structured rationales for model rankings, explicitly referencing specific turn numbers.
• Ability to provide constructive feedback and detailed annotations.

• Experience in data annotation, AI quality evaluation, content moderation, or a related role.

RequiredAlso useful

Also listed: attention to detail, written communication.

Project and contracting process: Turing. Applications continue on the provider's website.