You will test how well AI agents handle everyday tasks on your smartphone. You will compare two agents side by side, guide them through important decisions, and document the results.
Each testing run takes about 1.5 hours, including screen recording, evidence collection, and scoring. Tasks may include bill payments, returns, email follow-ups, bookings, shopping, travel, family logistics, outings, and forms.
- Run realistic everyday scenarios with two AI agents.
- Approve or stop agents at important decision points.
- Record each attempt and collect supporting evidence.
- Score completion, quality, control issues, satisfaction, and time.
- Write clear notes about what worked, what did not, and how the experience felt.