Data Annotation Specialist
HireArt is helping our client find a Data Annotation Specialists to help clear a high-volume data backlog and support the company’s computer vision pipeline. In this role, you’ll focus on labeling large volumes of image and video data using internal annotation tools. By shifting labeling work from engineering teams to dedicated specialists, this role helps accelerate model development cycles and improve the quality of training data used in computer vision systems. The ideal candidate is detail-oriented, comfortable working with visual data for extended periods, and able to maintain both speed and accuracy in a high-volume labeling environment. Schedule: 20-30 hours per week until the current backlog is cleared. Hours may be adjusted afterward based on project needs. As a Data Annotation Specialist, you'll:
- Perform high-volume image and video labeling using internal proprietary annotation tools.
- Identify and flag model failure cases across multiple sensor and data types.
- Maintain strong labeling accuracy while meeting throughput targets.
- Track progress and ensure consistent performance across large datasets.
- Follow established labeling guidelines and escalate ambiguous cases when needed.
- U.S. citizenship is mandatory for CUI eligibility
- Experience with computer vision labeling, robotics annotation, or video analysis
- Strong attention to detail and ability to maintain accuracy in repetitive visual tasks
- Ability to work directly within internal software tools (no third-party labeling platforms)
- Ability to sustain focus and productivity across high-volume annotation workloads
Commitment: This is a part-time contract position staffed via HireArt with a 3-month minimum commitment (reassessed at 6 months). The role is remote and available to U.S.-based candidates. HireArt values diversity and is an Equal Opportunity Employer. We are interested in every qualified candidate who is eligible to work in the United States. Unfortunately, we are not able to sponsor visas or employ corp-to-corp.