Sr Principal Applied Scientist, Amazon Robotics

Do you want to create worldwide impact in robotics while solving challenges at the edge of robotics research? Our team in Amazon Robotics builds high-performance, real-time robotic systems that can perceive, learn, and act intelligently alongside humans—at Amazon scale. Our mission is to enable robots to interact safely, efficiently, and fluently with the clutter and uncertainty of real-world fulfillment centers. Our AI solutions enable robots to learn from their own experiences, from each other, and from humans to build intelligence that feeds itself.

We hire and develop subject matter experts in AI with a focus on computer vision, deep learning, intelligent control, semi-supervised and unsupervised learning. We target high-impact algorithmic unlocks in areas such as scene and activity understanding, simultaneous localization and mapping, closed-loop control, robotic grasping and manipulation—all of which have high-value impact for our current and future fulfillment networks.

We are seeking an exceptional Principal Applied Scientist to join our organization, who possesses a deep understanding of both classical computer vision techniques and state-of-the-art machine learning approaches. The ideal candidate will be hands-on, coding-proficient, and a thought leader capable of influencing the direction of ML applied to Perception and Robotics.

Key job responsibilities
* Set the technical direction for the group, bridging classical computer vision techniques with the latest deep learning approaches
* Drive research and development of advanced perception systems, incorporating the latest advancements in self-supervised and multi-modal learning
* Collaborate with top-notch scientists and engineers to deliver the world's most scalable and robust robotic systems
* Lead the integration of vision-language models and foundation models into perception pipelines
* Develop novel approaches for few-shot learning and domain adaptation in robotic perception
* Advance the state-of-the-art in areas such as 3D scene understanding, object pose estimation, and activity recognition
* Drive ideas to products using a comprehensive toolset including deep learning, active learning, and computer vision (camera selection, camera calibration, instance and semantic segmentation, pose estimation, activity understanding)
* Mentor and guide team members in adopting best practices and staying current with rapid advancements in the field