Leads the team you would join
Leads the work day to day
Owns the opening
From your own Gmail and LinkedIn.
About the role
This is a senior, hybrid, full-time research and engineering role in San Francisco, CA, for an engineer with six or more years of SLAM or related experience. It suits people who work across robotics, computer vision, and probabilistic estimation and want to bring research into production systems. The work supports AMD's world modeling stack, where mapping and localization are foundational. You'll design and advance localization and mapping systems that combine classical geometry with learning-based methods.
What you’ll do
- Design and implement SLAM systems using visual, visual-inertial, lidar, or multi-sensor setups.
- Build localization and mapping pipelines covering pose estimation, map management, loop closure, and global optimization.
- Prototype learning-based and hybrid SLAM methods that combine geometry with machine learning.
- Build scalable state estimation using factor graphs, filtering, and smoothing.
- Develop sensor fusion across cameras, IMUs, depth sensors, and lidar.
- Analyze failure modes such as perceptual aliasing, dynamic scenes, and drift.
- Create benchmarks and metrics for accuracy, robustness, and performance.
- Optimize real-time performance, memory use, and compute efficiency.
- Work with reconstruction, simulation, and infrastructure teams on downstream integration.
- Propose research directions, mentor teammates, and help set best practices.
What they’re looking for
- Six or more years of experience in SLAM, state estimation, robotics perception, or related areas.
- Strong grounding in probabilistic estimation, optimization, and geometric vision.
- Deep experience with at least one SLAM approach: visual, visual-inertial, lidar, multi-sensor, or hybrid.
- Proficiency in Python and/or C++ with research or production SLAM systems.
- Experience with numerical optimization libraries or robotics frameworks.
- Familiarity with learning-based perception and representation learning.
- Understanding of sensor characteristics, calibration, synchronization, and noise.
- A record of driving projects from concept to deployment in ambiguous settings.
What’s on offer
- Hybrid working arrangement in San Francisco, CA.
Skills
Questions about this role
How many years of experience are required?
Six or more years in SLAM, state estimation, robotics perception, or related areas.
Is the role hybrid or remote?
The role is hybrid and based in San Francisco, CA.
Is the salary listed?
No, the posting does not list a salary.
Which languages are used?
Python and/or C++.