Amazon
IND, TS, Virtual
The GO-AI Operations team is building the future of workplace safety at Amazon Robotics Fulfillment centers through an advanced AI-powered human detection system rated at SIL2 (Safety Integrity Level 2), achieving up to 99% accuracy in identifying human presence. The team works at the intersection of computer vision and operational safety, processing multi-modal image data — including RGB, thermal, and grayscale feeds — to train and validate ML models that protect associates on the fulfillment floor. As an Associate in this role, you will perform high-quality 2D image annotation, accurately segmenting humans, robots, and carts within defined Regions of Interest (ROI). Your work directly enables the ML pipeline that powers real-time safety decisions, making this a high-impact, metrics-driven operational role where precision and consistency are critical. If you thrive in structured environments, have a sharp eye for detail, and want your work to directly contribute to associate safety at scale, this is the role for you
Key job responsibilities
Perform accurate 2D segmentation and annotation of humans, robots, and carts across RGB and thermal images using annotation tools (e.g., CVAT, SageMaker Ground Truth)
Identify and annotate objects within the defined Region of Interest (ROI), while applying masks outside the ROI as per guidelines
Interpret multi-feed image data (RGB, thermal, grayscale) with high precision, distinguishing between overlapping or partially occluded objects
Maintain high accuracy and productivity metrics aligned with program expectations and weekly targets
Follow standard operating procedures (SOPs) and quality guidelines consistently across all annotation tasks
Review and validate annotations to ensure defect-free outputs, flagging edge cases or ambiguous frames for escalation
Quickly adapt to feedback from quality audits and apply learnings to improve annotation quality in subsequent batches
Participate in training and nesting phases, achieving ramp-up within defined timelines and demonstrating readiness through qualification assessments
Maintain data confidentiality and adhere to security protocols when handling image and video data from fulfillment centers
Collaborate with team leads and quality analysts to support continuous improvement in annotation accuracy and throughput
A day in the life
Associates work in 24x7 environment with rotational shifts. Associates would be working in a 9 hour shift, including pre-scheduled breaks. The shift timings would be subject to change every 3-4 months or as per business requirement. In case associate is working in night shift, night shift allowance will be provided as per applicable Amazon’s work policy.
Weekly Offs: Rotational two-consecutive day off (it is a 5-day working week with 2 consecutive days off, not necessarily Saturday and Sunday) or as per business discretion.
About the team
The GO-AI Operations team is part of Amazon's Global Operations AI organization, enabling automation and workplace safety at Amazon Robotics Fulfillment centers. The team supports an advanced SIL2-rated human detection system achieving up to 99% accuracy in identifying human presence within robotic work cells. By processing multi-modal image data — RGB, thermal, and grayscale — the team performs high-quality 2D segmentation and annotation of humans, robots, and carts, directly powering the ML models that drive safe, hands-free automation. The team operates in a 24x7, metrics-driven environment where precision and consistency are critical.
Our inclusive culture empowers Amazonians to deliver the best results for our customers. If you have a disability and need a workplace accommodation or adjustment during the application and hiring process, including support for the interview or onboarding process, please visit https://amazon.jobs/content/en/how-we-hire/accommodations for more information. If the country/region you’re applying in isn’t listed, please contact your Recruiting Partner.