Computer Vision Annotator

About the role

We are looking for a remote Computer Vision Annotator to create high‑quality labeled datasets that power computer‑vision and machine‑learning models. You will work with image and video streams, applying precise annotation techniques to support object detection, classification, segmentation, tracking and 3D perception tasks.

Key responsibilities

• Draw accurate bounding boxes, polygons, polylines and 3D cuboids around objects of interest.
• Place key points for pose estimation and landmark detection.
• Perform semantic and instance segmentation, labeling every pixel in an image.
• Annotate point‑cloud and LiDAR data for autonomous‑vehicle and robotics applications.
• Interpret complex labeling taxonomies and refine guidelines in collaboration with project managers and data scientists.
• Provide feedback on annotation tools (CVAT, Label Studio, 3D Slicer, Amazon SageMaker Ground Truth) to improve workflow efficiency.
• Identify edge cases, potential biases and report findings to enhance model robustness.
• Assist with basic data preparation tasks such as image resizing and format conversion.
• Track key performance indicators like throughput, precision and recall.

Required profile

• Minimum 2 years of experience in data annotation or quality assurance for computer‑vision projects.
• Strong understanding of computer‑vision concepts (object detection, image classification, point clouds, LiDAR).
• Excellent attention to detail and commitment to consistent, high‑quality output.
• Effective written and verbal communication in English.
• Ability to document workflows clearly and collaborate with cross‑functional teams.

Required skills

• Proficiency with annotation tools: CVAT, Label Studio, 3D Slicer, Supervisely, Amazon SageMaker Ground Truth.
• Experience with annotation techniques: bounding boxes, polygons, key‑point labeling, 3D cuboid creation, semantic/instance segmentation.
• Familiarity with point‑cloud and LiDAR annotation.
• Knowledge of quality‑control processes such as consensus labeling, audit sampling, gold‑task seeding and inter‑annotator agreement metrics.