Senior Applied Scientist, Edge Vision for Robotics

Amazon Science; Massachusetts; On-site; USD 180,000 - 280,000; Full time

Benefits offered by this job

Health insurance

401(k) matching

Paid time off

Job summary

Amazon Robotics is seeking a Senior Applied Scientist focused on computer vision and robotics. You will shape perception systems for production edge hardware, influence sensor and compute architecture, and lead ML experimentation that informs hardware decisions across cameras, depth sensors, and edge modules.

You will work closely with hardware and software teams to deliver production-grade perception and sensing capabilities, contribute to roadmap planning, and push the boundaries of robotic

Qualifications

• Strong background in ML and computer vision for robotics.
• Experience with edge devices and sensor hardware.
• Ability to design and evaluate perception systems for production hardware.

Responsibilities

• Partner with hardware and science leads to define sensing and compute hardware requirements.
• Develop and optimize CV and ML models for robotic perception for edge deployment.
• Lead ML experimentation and validation to inform hardware architecture decisions.
• Design benchmarking frameworks to evaluate sensing hardware performance across use cases.
• Optimize models for production edge deployment including quantization, pruning, and distillation.
• Drive research on multi-modal sensor fusion, safe vision, and real-time inference for autonomous robotics.
• Collaborate with hardware, software, and program teams to transition models from prototype to production.

Skills

Machine learning

Computer vision

Robotics

Edge compute

Hardware collaboration

Job description

Amazon Robotics is seeking a Senior Applied Scientist focused on computer vision and robotics. You will shape perception systems for production edge hardware, influence sensor and compute architecture, and lead ML experimentation that informs hardware decisions across cameras, depth sensors, and edge modules.

You will work closely with hardware and software teams to deliver production-grade perception and sensing capabilities, contribute to roadmap planning, and push the boundaries of robotic