Applied Scientist, Perception: Real-Time VIO/SLAM

Amazon Inc.; Massachusetts; On-site; USD 150,000 - 210,000; Full time

Job summary

Amazon Robotics in New York seeks an Applied Scientist to develop and optimize Visual Inertial Odometry (VIO) and sensor fusion for our intelligent robots. You will own the full pipeline from algorithm design to embedded deployment on resource-constrained hardware, delivering real-time state estimation.

You will collaborate with hardware, controls, and navigation teams, apply ML-based perception to enhance classical methods, and lead projects from prototyping to production in a hands-on robotics

Qualifications

• PhD or Master + 4+ years in CS/CE/ML or related field.
• Hands-on experience deploying VIO/SLAM on robots.
• Experience with multi-sensor fusion and state estimation.
• Proficiency in C++, Python, and modern ML approaches.
• Experience with embedded hardware and real sensor data.

Responsibilities

• Design and implement VIO algorithms for real-time state estimation.
• Develop multi-sensor fusion pipelines for pose tracking.
• Optimize perception and tracking for embedded hardware under latency/power limits.
• Apply ML-based perception to complement geometric methods.
• Build calibration/evaluation/benchmarking infrastructure.
• Collaborate across hardware, controls, and navigation teams.
• Lead projects from research to production deployment.

Skills

VIO / SLAM

Sensor fusion

Embedded systems

C++

Python

State estimation

Multi-sensor fusion

Education

PhD in CS/CE/ML or related

Master's degree in CS/CE/ML or related

Tools

ROS/ROS2

NVIDIA Jetson / embedded platforms

ARM / GPU-accelerated devices

Job description

Amazon Robotics in New York seeks an Applied Scientist to develop and optimize Visual Inertial Odometry (VIO) and sensor fusion for our intelligent robots. You will own the full pipeline from algorithm design to embedded deployment on resource-constrained hardware, delivering real-time state estimation.

You will collaborate with hardware, controls, and navigation teams, apply ML-based perception to enhance classical methods, and lead projects from prototyping to production in a hands-on robotics