Postdoctoral Researcher – Visual Localization & Navigation | Last Mile

Description

Amazon is looking for talented Postdoctoral Scientists to join our research team for a full-time research position focused on visual localization and navigation for real-world applications.

Our work focuses on developing next-generation assistive technologies and logistics platforms that rely on robust, scalable visual perception systems. We are building solutions that enable devices and agents to understand, localize within, and navigate complex real-world environments—from indoor spaces with dynamic layouts to large-scale outdoor settings.

We are looking for Postdoctoral Scientists to work at the intersection of computer vision, SLAM, and scene understanding—supporting innovations that will be deployed to real systems at global scale.

The core technical challenges include building metric-semantic maps of complex environments, performing robust visual relocalization under appearance change, maintaining long-term map consistency, and achieving accurate monocular localization using both geometric and learning-based approaches—all under real-time constraints on real hardware.

The solution space is deliberately open-ended. We are looking for researchers who want to push the boundaries of visual localization and spatial AI—and see their work running on real platforms within months.

• Work closely with a senior science advisor, collaborate with other scientists and engineers, and be part of Amazon’s vibrant and diverse global science community.
• Publish your innovation in top-tier academic venues and hone your presentation skills.
• Be inspired by challenges and opportunities to invent cutting-edge techniques in your area(s) of expertise.

• PhD in computer vision, robotics, SLAM, or a related field received within 2 years of starting the program.
• Proven publication record in visual localization, SLAM, 3D scene understanding, or related technical fields.
• Experience in metric-semantic mapping, visual relocalization, or monocular localization.
• Proficiency in technologies relevant to the subfield (e.g., deep learning frameworks, geometric optimization, 3D reconstruction).
• Speak, write, and read fluently in English.

• Contributions to open-source software (e.g., MIT-licensed projects on Git Hub) demonstrating practical system-building skills alongside strong research output.
• Experience with metric-semantic mapping, object-centric mapping, or semantic SLAM systems.
• Familiarity with geometric and learning-based approaches to monocular localization and geometrically consistent relocalization.
• Experience with loop closure detection, long-term map consistency, and place recognition.
• Strong programming skills in Python and C++; familiarity with frameworks like PyTorch, ROS, Open3D, or GTSAM.
• Experience of making research work in practice—turning ideas into implementations that run on real systems, not just in simulation or on paper.
• Experience designing experiments and performing rigorous statistical analysis of results.

Ability to make principled tradeoffs between model accuracy, inference speed, and system constraints.

Amazon is an equal opportunity employer and does not discriminate on the basis of protected veteran status, disability, or other legally protected status.

Los Angeles County applicants:
Job duties for this position include: work safely and cooperatively with other employees, supervisors, and staff; adhere to standards of excellence despite stressful conditions; communicate effectively and respectfully with employees, supervisors, and staff to ensure exceptional customer service; and follow all federal, state, and local laws and Company policies. Criminal history may have a direct, adverse, and negative relationship with some of the…