We are an Australian deep-tech startup building \"Human Telemetry\" and infrastructure inspection systems. We are looking for an experienced Data labeller to manage our computer vision datasets. You will be the \"Human-in-the-Loop,\" ensuring our AI models are trained on perfect data. When not working on data, you will assist the founders with technical operations and hardware sourcing.
Remote Work - (Must overlap 4+ hours with Sydney, Australia Time - AEDT)
Interested Applicants send your RESUME/CV TO
Core Responsibilities (70% - Data & AI)
* Skeleton/Pose Estimation: Manually annotating human joints (key points) on video frames to train skeletal tracking models (e.g., YOLO Pose).
* Bounding Boxes & Polygons: Drawing tight, accurate bounding boxes around vehicles, structural components, and safety hazards.
* Quality Assurance (QA): Reviewing the output of our automated AI models to correct false positives or \"drift\" (e.g., fixing where the AI missed a person's arm or leg).
* Dataset Management: Organizing raw video footage from field trials (terabytes of data), snipping relevant clips, and uploading them to labelling platforms (CVAT, Roboflow, or Label Studio).
Secondary Responsibilities (30% - Ops & Admin)
* Hardware Sourcing: Finding and identifying stock levels for specific components (e.g., NVIDIA Jetson modules, industrial cameras, server racks) from global suppliers.
* Documentation: helping draft technical manuals or simple \"How-To\" guides for our installation partners.
#J-18808-Ljbffr