The Work
You will test two AI agents side by side on realistic everyday errands. You will use your phone and personal accounts, guide each agent when needed, and observe how well it completes each task.
You will screen record every attempt and explain what worked, what failed, and how the experience felt. Your feedback will cover completion, quality, user control, satisfaction, and time.
- Run scenarios involving bills, returns, email follow-ups, bookings, shopping, groceries, travel, outings, forms, and other errands.
- Approve or stop an agent when an important action needs your control.
- Document each interaction clearly from start to finish.
- Score each run and provide precise written feedback about the agent experience.
What It Pays And Takes
This is remote, part-time contractor work for candidates based in the United States. You will complete the work with a smartphone, personal accounts, and the required testing tools.
- Pay: $25 to $45 per hour.
- Location: Open to candidates in the United States.
- Education: Bachelor's degree required.
- Experience: Two to three years of experience preferred, especially in engineering, business management, office, or operations work.
- Skills: Strong attention to detail, precise instruction following, clear writing, and careful step-by-step documentation.
- Background: You should use AI in daily life. Generalists are welcome, and no coding background is required.
- Equipment and accounts: A smartphone that can record its screen, an active personal AI assistant account on the Plus tier, and active access to the required testing tools.
- Testing purchases: A payment card is required, and test purchases are reimbursed afterward.