Senior robotics navigation engineer

Dexmate is building the foundation for physical AI — a unified platform that combines high-quality robotic hardware with a universal Physical AI OS, making robots as easy to build and deploy as software. Today, robotics is fragmented, slow, and closed: most builders are forced to reinvent the same stack again and again, and most ideas never make it past the prototype stage. We exist to change that. Our mission is to democratize robotics by lowering the barrier to entry, delivering a plug-and-play platform for developers, researchers, and enterprises, and cultivating an open ecosystem that accelerates the evolution of physical AI. If you want to help shape the next layer of human capability — and believe the future of robotics should be built together, not in isolation — we'd love to build it with you.

The Role

We are looking for a Senior Robotics Navigation Engineer to own the localization, mapping, and navigation stack for our humanoid robots. You will design and implement 3D SLAM pipelines, multi-modal state estimation systems, and real-time navigation algorithms that enable our robots to understand where they are, build accurate maps of their environment, and move through it reliably.

This is a production-focused role. You are not prototyping algorithms in simulation — you are deploying them on hardware, validating them in real environments, and owning their reliability at scale. We want engineers who have closed that loop before: on self-driving cars, AGVs, mobile robots, or similar deployed autonomous systems.

Responsibilities

Design, implement, and deploy production-grade 3D SLAM and localization systems fusing data from LiDAR, RGB-D cameras, IMUs, wheel encoders, and proprioceptive signals

Build and maintain state estimation pipelines — Kalman Filters, Extended Kalman Filters (EKF), Unscented Kalman Filters (UKF), or Factor Graph backends (GTSAM, Ceres, g2o) — with reliable accuracy in dynamic, GPS-denied, and perceptually degraded environments

Develop real-time 3D navigation algorithms: costmap generation from point clouds, global and local path planners (sampling-based, optimization-based), and traversability analysis

Implement sensor calibration pipelines (intrinsic and extrinsic) for multi-sensor rigs; own the calibration quality that underpins system accuracy

Design and build the evaluation and regression frameworks that prove the navigation stack is working correctly — logging, metrics, replay tooling, and failure analysis infrastructure

Collaborate with perception, controls, and hardware teams to integrate the navigation stack end-to-end into the full robot autonomy system

Troubleshoot and resolve challenging field failures; own root cause analysis and system-level fixes when localization or navigation breaks in deployment

Mentor junior engineers and contribute to technical roadmap planning for the autonomy stack

Minimum Qualifications

5+ years of industry experience in robotics autonomy, with a primary focus on SLAM, localization, or state estimation

Proven track record of deploying navigation or SLAM systems on real autonomous platforms — self-driving vehicles, AGVs, mobile robots, or equivalent — not just simulation or research prototypes

Deep theoretical and practical command of probabilistic robotics: Bayesian filtering, sensor fusion, covariance modeling, and non-linear optimization

Hands-on experience with 3D SLAM modalities: LiDAR SLAM, Visual SLAM (VSLAM), Visual-Inertial Odometry (VIO), or multi-modal fusion

Proficiency in C++ (C++14/17 or newer) for real-time, performance-critical code; strong software engineering fundamentals

Experience with optimization libraries: GTSAM, Ceres Solver, g2o, or equivalent factor graph backends

Familiarity with ROS/ROS 2 and standard robotics tooling

Ability to explain why a localization module failed in a specific scenario to both a technical peer and a non-technical stakeholder

Preferred Qualifications

M.S. or Ph.D. in Robotics, Computer Science, Electrical Engineering, or related field

Experience with learning-based or hybrid approaches to localization and mapping (e.g., neural implicit maps, foundation model-assisted SLAM)

Background in semantic SLAM or scene understanding — associating geometric maps with object-level semantics

Experience with GPU acceleration (CUDA) for perception or navigation pipelines

GNSS/RTK fusion experience for outdoor or GPS-blended deployments

Familiarity with map management at scale: map storage, versioning, sharing across a robot fleet, and lifecycle management

Prior experience in a startup or fast-moving R&D environment with an emphasis on shipping

Contributions to open-source SLAM or navigation frameworks (ORB-SLAM, RTAB-Map, Cartographer, LIO-SAM, KISS-ICP, etc.)