AI Agent Tester for Real-World Task Evaluation
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.
- Field
- Generalist
- Where
- United States
- Commitment
- part time
- Posted
- 5 days ago
The full listing
As posted on OpenTrain AI. See the original
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.
How It Works
Apply on OpenTrain with your resume, then complete the application on the hiring site.
About AI Training Work
AI training is the human work behind artificial intelligence. People test systems, review their outputs, and explain what needs to improve so AI tools work better in real situations.
How OpenTrain AI pays and hires
OpenTrain says it is free for freelancers, with no fees or subscriptions. Clients fund each milestone before work starts, and OpenTrain pays out to your connected payout method when the client approves the milestone. It doesn't publish a fixed payout schedule.
Create a free profile with your domains, languages and hourly rate, then apply to listings. Some listings hire through OpenTrain; others link you straight to the hiring platform.
Source: OpenTrain: AI training careers, OpenTrain docs: invoicing and payouts
Full OpenTrain AI profile: watch-outs and worker reports →Other generalist roles
- Luxury Automotive Strategy ExpertMercor$300-350/hr
- Intelligence Expert (Bachelor's, Master's and PhDs) - AI TrainerHandshake AIUp to $200/hr
- IntensivistHandshake AIUp to $180/hr
- AI Written Response Evaluator and AssessorOpenTrain$90-140/hr
- Cardiac Sonographer (RDCS)Handshake AIUp to $130/hr
- General Sonographer (RDMS)Handshake AIUp to $130/hr