Technical Lead - State Estimation

Overland AI•Seattle, WA

About The Position

Overland AI is building autonomous ground vehicles capable of operating where GPS is unreliable, terrain is unpredictable, and failure is not an option. We're looking for a Technical Lead in State Estimation to define the algorithms that allow our vehicles to understand exactly where they are—and keep them operating confidently in the world's most demanding environments. In this role, you'll lead the architecture and development of our state estimation stack, solving challenging problems across localization, mapping, sensor fusion, and probabilistic inference. You'll work at the intersection of cutting-edge robotics research and production autonomy, turning advanced estimation techniques into robust systems that perform reliably in real-world deployment. As a technical leader, you'll set the direction for state estimation at Overland, mentor a team of exceptional robotics engineers, and collaborate across perception, planning, controls, and platform engineering to build the next generation of autonomous off-road vehicles.

Requirements

MS or PhD in Robotics, Computer Science, Electrical Engineering, or a related technical field with specialization in state estimation, SLAM, localization, or probabilistic inference.
5+ years developing production-grade state estimation or SLAM systems deployed on physical robotic platforms.
Deep expertise in probability theory, Bayesian estimation, optimization, and nonlinear inference, including: EKF, UKF, Error-State Kalman Filters
Factor graph optimization (GTSAM, Ceres, g2o)
MAP/MLE estimation
Demonstrated experience deploying robust estimation systems in complex, unstructured, or off-road environments.
Expert-level C++ and strong Python development skills.
Experience building low-latency, real-time robotics software.
Strong understanding of inertial navigation, sensor calibration, and multi-modal sensor fusion (IMU, LiDAR, cameras, GNSS).
Publications or significant technical contributions in SLAM, visual-inertial odometry, LiDAR-inertial odometry, localization, or related fields.

Nice To Haves

PhD in Robotics, Computer Science, Electrical Engineering, or a related field.
Experience incorporating machine learning into estimation pipelines.
Experience with ROS 2 and real-time middleware (DDS, shared-memory transport).
Experience with terrain-relative navigation, prior-map localization, or GPS-denied navigation.
Contributions to widely used open-source robotics or estimation libraries.
Experience leading technical teams, mentoring engineers, and driving architecture decisions.
Experience shipping autonomy systems on production robotic or autonomous vehicle platforms.

Responsibilities

Design and implement odometry, localization, and mapping algorithms that enable reliable autonomy in GPS-denied and degraded environments.
Develop robust multi-sensor fusion systems combining IMUs, LiDAR, cameras, GNSS, wheel encoders, and other onboard sensors.
Formulate and solve estimation problems using Kalman filtering, Bayesian inference, factor graphs, nonlinear optimization, and modern probabilistic techniques.
Evaluate and integrate learned approaches—including learned odometry, feature representations, and neural mapping methods—where they deliver measurable improvements over classical techniques.
Develop high-performance, production-quality C++ (C++23) software optimized for real-time robotic systems.
Build tooling, simulation infrastructure, and evaluation pipelines that enable rapid algorithm development and validation using large-scale field datasets.
Lead verification and validation efforts across diverse terrain, weather conditions, and operational environments.
Partner closely with perception, planning, controls, and systems engineers to deliver an integrated, reliable autonomy stack.
Mentor engineers, drive technical excellence, and establish best practices for robotics software development.