Applied Scientist - Computer Vision, Amazon Robotics

About the role

At Amazon Robotics we build 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 in high-density real-world fulfillment centers. We invent and scale AI systems for robotics in fulfillment, targeting high-impact algorithmic unlocks in scene and activity understanding, large-scale generative models, closed-loop control, robotic grasping and manipulation. As an Applied Scientist, you will drive ideas to products using deep learning, semi-supervised learning, and active learning, while leading and mentoring a team of applied scientists and engineers.

Responsibilities

• Architect, design, and implement machine learning models for vision systems on robotic platforms
• Optimize, deploy, and support at scale ML models on the edge
• Influence the team’s strategy and contribute to long-term vision and roadmap
• Work with stakeholders across science, engineering, and operations to iterate on design and implementation
• Maintain high standards by participating in reviews, designing for fault tolerance and operational excellence, and creating mechanisms for continuous improvement
• Prototype and test concepts or features through simulation, emulators, and live robotic equipment
• Work directly with customers and partners to test prototypes and incorporate feedback
• Mentor other engineering team members

About the Team

The Vulcan Stow Perception team builds the visual intelligence that enables Amazon’s next-generation robotic stow systems to understand and interact with densely packed fulfillment environments. We own the full perception stack—from raw sensor input to actionable 3D scene representations—powering robots that autonomously stow millions of items daily across Amazon’s global network. We tackle unsolved problems in 3D robotic perception: completing occluded scenes, generating real-time semantic occupancy predictions, fusing multi-camera inputs, and producing sub-250ms mesh reconstructions that drive downstream manipulation decisions. We operate at the intersection of research and production-scale deployment, building systems that are both scientifically rigorous and operationally robust.

Basic Qualifications

• 3+ years of experience building models for business applications
• PhD, or Master’s degree and 4+ years of experience in CS, CE, ML, or related field
• Experience developing and implementing deep learning algorithms, particularly in computer vision
• Experience with patents or publications at top-tier peer-reviewed conferences or journals
• Programming experience in Java, C++, Python, or a related language

Preferred Qualifications

• Experience analyzing large volumes of data to drive decisions
• Knowledge of statistical modeling
• Experience in a fast-paced, dynamic organization
• Strong communication skills with leadership, stakeholders, and cross-functional teams
• Extensive experience writing scientific papers with multiple publications in top ML, computer vision, or robotics conferences/journals (e.g., CVPR, ICCV, ECCV, PAMI, NeurIPS, IROS, CoRL)

Pay

USA, WA, Seattle — $142,800.00 – $193,200.00 USD annually (base salary range). Final compensation includes sign-on payments and restricted stock units (RSUs), and is determined based on experience, qualifications, and location.