In this role, you'll use your everyday experience managing your digital life — email, calendars, bookings, and coordination — to help train and evaluate next-generation AI assistants. Your work will shape how AI models handle the real tasks people delegate every day.

• Complete realistic, everyday digital tasks (emailing, scheduling, booking, coordinating) using AI assistant tools under standardized, repeatable conditions.
• Give an AI assistant clear instructions for real-world tasks, then carefully review and verify its work for accuracy and completeness.
• Apply detailed grading guidelines to evaluate, rank, and annotate assistant responses, providing structured feedback that improves model performance.
• Test scenarios involving email threads, calendar invites, online bookings, and multi-person coordination, ensuring coverage of diverse situations and edge cases.
• Utilize AI training tools and platforms to record findings and submit evaluation data in alignment with project standards and milestones.
• Collaborate with project trainers and contributors to resolve ambiguities, refine guidelines, and ensure evaluation consistency across the team.
• Participate in ongoing quality reviews, incorporating feedback to maintain rigorous evaluation standards.

• Prior experience with AI evaluation, data labeling, user testing, or similar feedback-driven projects.
• Experience grading or reviewing AI outputs for accuracy, helpfulness, and safety.

• Are based in the US, have an iPhone with iMessage, and have a Facebook or Instagram account.

Work arrangement: Fully remote

Eligible countries: United States