Amazon Robotics is transforming warehouse automation through edge AI and machine learning applied to real-world robotics challenges. We're seeking a Senior Applied Scientist to advance our mobile manipulation capabilities by developing learning-based approaches that enable robots to navigate and manipulate objects in dynamic fulfillment environments.
Responsibilities:
- Design, develop, train, and deploy deep learning models for perception tasks (e.g., object detection, segmentation, pose estimation, tracking)
- Develop and maintain robust camera calibration pipelines (intrinsic, extrinsic, hand-eye calibration, multi-camera systems)
- Build perception systems for robotic manipulation including grasp detection, object pose estimation, and visual servoing
- Improve model performance through architecture optimization, data curation, and training strategies
- Build and maintain production-quality perception codebases with proper testing and documentation
- Profile and optimize models for real-time inference on embedded/edge platforms
- Collaborate with cross-functional teams (robotics, motion planning, controls) to integrate perception outputs for manipulation tasks
- Establish best practices for model versioning, experiment tracking, and MLOps
- Mentor junior engineers and contribute to technical roadmap planning