About UsFax Research produces high-quality video, multimodal, and task-demonstration datasets for robotics teams building humanoids, service robots, and warehouse automation systems. Our datasets train next-generation robotics AI, and accurate, high-quality annotation is central to what makes our data valuable.

About the RoleWe're looking for a detail-oriented Data Labeller to annotate real-world task-demonstration video footage; labelling actions, objects, and movements used to train robotics models. This is a remote contract role, ideal for someone with strong attention to detail, prior annotation experience, and a good eye for consistency across large volumes of footage. Location not limited to Australia.

Responsibilities

- Annotate video footage using bounding boxes, polygon/segmentation, keypoints, and temporal action labels
- Apply detailed labelling guidelines consistently across large batches of footage
- Flag ambiguous, unclear, or edge-case footage to the team lead for clarification
- Meet quality benchmarks (inter-annotator agreement, accuracy targets) and turnaround deadlines
- Participate in a paid trial/assessment task as part of the hiring process
- Provide feedback on labelling guidelines to help improve clarity and consistency over time

Requirements

- Bachelor's degree (Computer Science,



IT, or related field) preferred
- 1-2+ years of experience in data annotation, data entry, QA, or transcription (annotation tool experience is a strong plus; e.g. CVAT, Labelbox, Scale AI, VoTT)
- Strong attention to detail and consistency across repetitive, high-volume tasks
- Basic computer literacy (file management, spreadsheets, ticketing/task-tracking systems)
- Solid written English proficiency
- Ability to work hours that overlap with Australian business hours (AEST/AEDT) for at least part of the day
- Reliable internet connection and a quiet remote work setup, work from anywhere, location not limited to Australia

Nice-to-Haves
- Experience with video-specific annotation (frame-by-frame tagging, temporal action segmentation)
- Familiarity with basic robotics or automation concepts
- Background in gaming QA, film/video editing, or transcription

How to ApplyPlease submit your resume and sample work (if any) along with a short response to: "Describe a time you had to maintain high accuracy on a repetitive task" to [email protected].

Shortlisted candidates will be invited to complete a short, paid sample annotation task as part of the assessment process.

📌 Data Labeller (Australia)
🏢 Fax Research
📍 Australia