Key skills
AICommunication
About this role
Role Overview
- Design and run short multi-turn conversations (typically 1–5 turns) intended to test AI personalization behavior
- Create prompts grounded in realistic personal scenarios to evaluate contextual understanding
- Review AI responses to determine whether personalization is correctly applied
- Check grounding quality to ensure the model does not invent unsupported claims about the user
- Evaluate integration quality — confirming personal signals are used naturally (not forced or robotic)
- Compare two responses side-by-side and determine which is more helpful, natural, and relevant
- Write clear, structured rationales explaining rankings and referencing specific conversation turns
- Verify debug information showing whether correct data sources were used
- Maintain strict workflow hygiene (including deleting evaluation conversations when required)
Requirements
- Strong Polish proficiency (reading & writing required) — Polish is the primary evaluation language
- BS/BA degree or equivalent experience in a relevant field (e.g., Policy, Law, Ethics, Linguistics, Journalism, Computer Science, or a related analytical field)
- Strong analytical thinking and ability to assess nuanced AI outputs
- Excellent written communication skills with ability to produce structured evaluation notes
- High attention to detail when comparing similar responses
- Ability to work independently in a fully remote environment
- Reliable desktop/laptop and stable internet connection
- Willingness to use your primary personal Google account and enable personal data sources for evaluation purposes