Zivney Robinson

Key Skills

Software

Data Annotation Tech

Top Subject Matter

No subject matter listed

Top Data Types

3D Sensor

Top Label Types

Polygon

Point Key Point

Polyline

Segmentation

Cuboid

Freelancer Overview

I am a detail-oriented Data Annotation Specialist with extensive experience in 3D LiDAR labeling, point cloud segmentation, and sensor-fusion workflows for autonomous systems. My expertise includes creating precise 3D bounding boxes, semantic and instance segmentation, and tracking objects across frames to ensure high-quality training data for computer vision models. I am skilled in using tools like CVAT, Supervisely, Scale, and various point cloud visualization platforms, and have consistently maintained accuracy rates above 90% while meeting tight deadlines. I thrive in collaborative environments, contribute to quality assurance, and am committed to continuously improving both productivity and data quality in every project I undertake.

Labeling Experience

Tasker

Data Annotation Tech3D SensorPolygonPoint Key Point

I supported a large-scale data preparation initiative focused on building high-quality training datasets for autonomous vehicle perception systems. The project transformed raw 3D LiDAR point clouds and synchronized sensor data into structured annotations for machine learning pipelines. My responsibilities included creating accurate 3D cuboids around vehicles, pedestrians, cyclists, and roadway assets; performing semantic and instance classification; and maintaining object identity through frame-to-frame tracking. The work covered extensive urban and highway scenarios and required consistent delivery against defined daily and weekly throughput targets across thousands of frames. Quality and consistency were central to the workflow. I adhered strictly to detailed taxonomies, client guidelines, and standard operating procedures while performing rigorous self-checks before submission. Datasets passed through multi-level validation pipelines involving peer review and supervisory audits.

I supported a large-scale data preparation initiative focused on building high-quality training datasets for autonomous vehicle perception systems. The project transformed raw 3D LiDAR point clouds and synchronized sensor data into structured annotations for machine learning pipelines. My responsibilities included creating accurate 3D cuboids around vehicles, pedestrians, cyclists, and roadway assets; performing semantic and instance classification; and maintaining object identity through frame-to-frame tracking. The work covered extensive urban and highway scenarios and required consistent delivery against defined daily and weekly throughput targets across thousands of frames. Quality and consistency were central to the workflow. I adhered strictly to detailed taxonomies, client guidelines, and standard operating procedures while performing rigorous self-checks before submission. Datasets passed through multi-level validation pipelines involving peer review and supervisory audits.

2018 - 2025

Education

Work History