Chores at home
Cooking, dishes, laundry and cleaning in your own home, filmed with a head camera while you do them.
Key facts
- Pay ranges from USD 5–15 per hour of accepted footage, depending on the project. It varies by project and is not guaranteed, and you see the exact rate before you accept.site FAQ; Data Collector Agreement §2.1
- Review usually takes about a week (contract limit 30 days). After acceptance you're usually paid within 7 days, in US dollars through Wise; the contract limit is 60 days.Data Collector Agreement §2.1, §3
- You don't need to speak English: instructions and support are available in Spanish and Portuguese.TaskCapture
- There is no fee to apply, and no coding or AI background is needed.TaskCapture
- You never pay for equipment: the project provides it, or the task needs no special equipment.TaskCapture
- You must be 18 or older, and every identifiable person on camera must sign a release.Privacy Policy §10; §2 "If you appear in footage"
- TaskCapture works with hundreds of data collectors in 50+ countries.homepage network block
Is there AI training work for people with no tech experience?
Yes. You don't need coding or AI experience. You record a task you already know how to do, such as cooking, stocking shelves or fixing a bike, from your own point of view, and teams use that footage to train robots and AI models.
A line cook already knows how a prep shift runs, and a gamer knows how a match plays out. You don't need to know how AI models get built. Each project lists its own requirements before you accept it.
How does your footage train AI?
An AI model learns from examples. To learn how people chop an onion, fold a towel or restock a shelf, it needs many recordings of real people doing that, in real places. Our explainer on what AI training data is covers it in more detail.
Footage filmed from the worker's own eyes is called egocentric, or first-person, video. It shows the hands, the tools and the order of the steps, and a robot has to copy all of that. Researchers have built large public collections of it: the Ego4D dataset gathered 3,670 hours of daily-life video from 931 camera wearers. Robotics researchers have also shown that video of people can help robots learn. The R3M study trained a robot's vision on Ego4D's human video and found it helped robots learn manipulation tasks from few demonstrations. Read more in what egocentric video is.
What kind of footage do collectors record?
TaskCapture collects three kinds of first-person footage. Each project says which kind it needs.
Gameplay recording works differently from physical work, and from apps that pay you to play. Our guide to getting paid to record gameplay explains the difference. Researchers use the same kind of data: OpenAI's Video PreTraining study trained an agent on gameplay video paired with keyboard and mouse actions.
Each activity, from a professional kitchen to farm work, has its own page among the types of recording work.
Shifts at work
Restaurant prep lines, warehouses, retail, workshops and trades, hotels and farms, filmed during a normal shift.
Gameplay, with the inputs
The game screen, recorded with every keyboard, mouse or controller input in sync. No camera, and no voice, chat or gamer handles.
Is it remote?
Partly. TaskCapture's residential projects happen in your own home, and gaming projects record your game screen and inputs. Commercial projects happen at your workplace, during a normal shift, and only with written permission from the owner or manager.
You can't do this from a café with a laptop. The footage has to show a real task in the place where it happens.
How does a project work?
TaskCapture arranges each project with you directly, by email. There is no account or app to set up today.
- 01
Apply
Fill in the short form below with your country and where you work.
- 02
We email you
When a project fits where you work, we send you its details: the task, the setting, the requirements, the equipment and the pay.
- 03
You accept
You decide. The pay is fixed once you accept.
- 04
You record
You record the task as the project describes, with the releases and permissions it needs.
- 05
You hand it over
The project details tell you how to send your footage to TaskCapture.
- 06
Quality review
People and automated checks review your footage. Review usually takes about a week, and the contract limit is 30 days. If something needs fixing, you get a revision request and 5 days to send a corrected version.
- 07
You get paid
After your footage is accepted, you're paid in US dollars through Wise, usually within 7 days and never later than 60 days after acceptance.
Read what you wear and how recording works for the camera setup and what a good recording shows.
A worked example: one commercial project, start to finish
Maya is an invented person. Every rule in her story comes from the steps above.
Maya is a line cook in the United States. She applies with her country and "restaurant kitchen". Later, when a project fits, she gets an email about it: record prep tasks during normal shifts. The email states the task, the camera to wear, the requirements and the rate per hour of accepted footage. She checks the rate and accepts, and from that moment the rate is fixed.
Before recording, she gets written permission from her kitchen manager and a signed release from the two coworkers who will be on camera. She records three prep sessions and sends the footage the way the project details describe. About a week later, review asks her to redo one clip because her hands left the frame. She sends a corrected clip within the 5 days allowed. Once the footage is accepted, she is usually paid within 7 days, in US dollars, and never later than 60 days after acceptance. She is paid for the accepted hours at the rate she accepted.
Recording work compared with other AI training jobs
"AI training job" covers very different work. Our guide to data collection compared with annotation jobs goes deeper.
| Recording your work (TaskCapture) | Labeling or rating data | Driving robots remotely (teleoperation) | |
|---|---|---|---|
| What you do | Record a real task you already do, from your point of view | Tag images, rate chatbot answers or write text at a computer | Control a robot through a task with special equipment |
| Where | Your home, your workplace (with permission) or your gaming setup | At a computer | Usually at a lab or a company site |
| What you need | The project's camera or none; no AI background | Depends on the platform | Training on the robot |
| Does TaskCapture offer it? | Yes | No | No |
Robotics teams use teleoperation data as well: the Open X-Embodiment project pooled robot data from 22 different robots. TaskCapture doesn't hire people to drive robots. Collectors record their own work.
How do consent and privacy work?
Everyone on camera agrees first. You must get a signed release from every identifiable person you record. In a commercial setting, such as a restaurant or warehouse, you also need written permission from the owner or manager. A person under 18 may appear only with a release signed by a parent or guardian; otherwise the footage is rejected and deleted.
TaskCapture does not run face or voice recognition. Footage goes to data buyers only after it passes quality review, datasets never include your name, contact details or payment data, and buyers are contractually barred from identifying anyone in the footage. The Privacy Policy and the Data Collector Agreement set out the rules in full. Before you record at work, read recording at work: privacy and consent.
Common recording mistakes to check against your project
Check your project's instructions first. For more recording tips, see what gets footage accepted.
- Recording without permission. At a workplace you need the owner's or manager's written permission first, and a release from every identifiable person on camera.
- Losing the task. Your hands or the work leave the frame, or the recording stops before the task is done. Check the project instructions for which steps and camera views to record.
- Filming private information: screens, documents or the faces of people who haven't signed a release.
- Recording before you accept. The project details set the task and the rate, so record only after you've accepted them.
- Missing a revision request. You have 5 days to send a corrected version, so watch your email after you hand footage over.
How to tell this from a job scam
Fake AI job offers are common, so check before you trust anyone. The US Federal Trade Commission warns that "honest employers, including the federal government, will never ask you to pay to get a job". The FTC also warns about task scams that show fake earnings in an app and then ask you to deposit your own money, usually in crypto, to get them out.
TaskCapture never asks for a fee to apply, and you never pay for equipment. Collectors record video; they don't rate chats or write answers. TaskCapture collects to order: a buyer makes a request, TaskCapture matches collectors who do those tasks, and they record for that project. Our checklist on how to spot AI training job scams works for any company, including us. The FTC's job scams guidance lists the warning signs.
Questions
How much does it pay?
Pay ranges from USD 5–15 per hour of accepted footage, depending on the project. It varies by project and is not guaranteed: you see the exact rate in the project details before you accept, and you can say no. Once you accept, the pay is fixed. You're paid in US dollars.
When do I get paid?
First your footage is reviewed, which usually takes about a week (the contract limit is 30 days). After it's accepted, you're usually paid within 7 days, in US dollars, through Wise. The Data Collector Agreement sets the limit: never later than 60 days after acceptance.
Do I get paid for rejected footage?
No. Rejected footage is not paid: pay is per hour of accepted footage. If review finds a problem it can fix, you get a revision request and 5 days to send a corrected version first.
Is this a real job or a gig?
It is project-based contract work. When you accept a project, you work on it as an independent contractor under the Data Collector Agreement, and the project details set out the task, the requirements and the pay. TaskCapture collects to order: a buyer makes a request, TaskCapture matches collectors who do those tasks, and they record for that project.
Do I need experience with AI, or my own camera?
No to both. No coding or AI background is needed, and you never pay for equipment: the project provides it, or the task needs no special equipment.
Can I record at my workplace?
Only with written permission from the site's owner or manager, and with a signed release from every identifiable person on camera. Footage without them may be rejected.
Who sees my footage?
People and automated checks at TaskCapture review it. Data buyers get it only after it passes review, without your name, contact details or payment data, and they are contractually barred from identifying anyone in it.
Which countries can apply?
TaskCapture works with hundreds of data collectors in 50+ countries. Tell us your country in the form; we email you when a project fits where you live and work. You don't need to speak English: instructions and support are available in Spanish and Portuguese.
To weigh us against other services, our recording-platform comparison asks every one of them the same questions.