You will test AI assistants using everyday email, calendar, booking, scheduling, and coordination tasks. You will check whether each assistant completes the task accurately and fully, then record your findings.
The work includes giving clear instructions, grading and ranking responses, adding annotations, and submitting structured evaluation data. You may also identify unusual cases, discuss unclear examples with other contributors and trainers, and use feedback during quality reviews.
- Complete repeatable tasks involving Gmail messages, email threads, calendar invitations, bookings, scheduling, and multi-person coordination.
- Check AI-generated work for accuracy, completeness, and adherence to evaluation guidelines.
- Grade, rank, and annotate assistant responses using structured AI training tools.
- Record findings and submit evaluation data.
- Identify edge cases and resolve unclear examples with contributors and trainers.