Long-term Visual Data Labeling - Images/Video - USA

About OpenTrain

OpenTrain is a centralized job board for data-labeling and AI-training work. We bring together projects from many AI companies and labeling platforms so contributors can find steady, relevant opportunities in one place.

Creating an OpenTrain account is free and applying takes only a few minutes. We connect skilled annotators with projects that need human expertise to make AI systems accurate and safe.

Why AI training (data labeling) matters

AI models learn from human-produced examples. Tasks like annotating images and video, drawing bounding boxes, labeling keypoints, and classifying frames provide the ground-truth data that computer-vision models require.

This work is often remote, flexible, and accessible: many projects allow you to set your own hours and contribute directly to how real-world AI systems behave.

The role

We are hiring experienced computer-vision data-labeling freelancers for a long-term, ongoing project focused on images and video. You will annotate visual data using bounding boxes, polygons, keypoints, cuboids, object-detection labels, and classification.

This is a contractor, part-time role. The client expects a steady commitment of approximately 15–20 hours per week, with flexibility to set your own schedule as long as tasks are completed on time.

• Work type: Contractor / Part-time
• Time commitment: ~15–20 hours per week (ongoing)
• Location requirement: Must be located in Canada or the USA
• Pay structure: Pay-per-hour at USD $14/hour

What you'll do day-to-day

Perform high-quality annotations on image and video datasets to prepare training data for computer-vision models. Tasks will vary and may include frame-by-frame labeling, multi-object tracking, and complex polygon segmentation.

Use annotation tools to create and verify labels, follow project guidelines, and deliver consistent, error-free work on deadlines.

• Draw bounding boxes, cuboids, and polygons for objects and regions of interest
• Place point keypoints for pose or landmark tasks
• Assign classification labels to images or video frames
• Perform object-detection and tracking across frames where required
• Review and QA your annotations to meet quality thresholds

Requirements

You must meet the experience and location requirements listed below; we will verify experience during the screening process.

All qualifications below come from the project brief and are required.

• 2–3+ years of hands-on image and video data labeling experience
• Proven experience with bounding boxes, classification, polygon annotations and other visual-label types
• Proficiency with annotation tools (examples: Labelbox, CVAT, VGG Image Annotator, or similar)
• High attention to detail and consistent, reliable output on long-term projects
• Strong communication skills for collaboration and reporting
• Ability to work independently and manage time effectively
• Location: Must be located in Canada or the USA
• Availability: Able to commit approximately 15–20 hours per week

Compensation and pricing notes

This project is paid hourly on a pay-per-hour basis at USD $14/hour as provided in the brief. The description also encouraged applicants to propose competitive rates considering the long-term nature of the work — please factor that into any discussions during screening.

You will be contracted as a freelancer and should be prepared to invoice or bill according to the platform or client process.

What we’ll ask in the interview

During screening we will focus on your annotation experience, tool familiarity, and examples of past work on image/video projects. Expect to discuss your process for quality control and how you manage deadlines on long-term contracts.

The client specifically wants candidates who can describe prior image/video annotation projects on major platforms.

• Describe 2–3 image/video labeling projects you've completed (focus on visual—text labeling not relevant)
• Which annotation tools have you used and what features do you rely on?
• Have you worked on platforms like Scale AI, Remotasks, Appen, or similar? Describe the project types and your role
• How do you ensure consistency and accuracy across long stretches of labeled data?

How to apply

If you meet the qualifications, apply via OpenTrain with a summary of your relevant image/video annotation experience, the tools you use, and your availability (hours per week). Include sample project descriptions or links to portfolios if available.

Because this is a long-term project, emphasize reliability and past experience on ongoing contracts. Applicants must be located in Canada or the USA.

• Submit your most competitive rate and confirm availability for ~15–20 hrs/week
• List annotation tools you’re proficient with and example projects
• Highlight experience on platforms like Scale AI, Remotasks, Appen if applicable