Senior Robotics Software Engineer - Localization

Samson Rose; Los Angeles (CA); On-site; USD 150,000 - 210,000; Full time

Job summary

Samson Rose in Los Angeles seeks a senior robotics software engineer to own the localization stack for our lunar rovers. You will design the approach, implement it in Python/C++, and test on hardware and Gazebo simulations using NASA terrain data.

You will define metrics for drift, accuracy, and health, validate results in field-like tests, and integrate localization with path planning and hazard avoidance to enable robust autonomous operation in GNSS-denied environments.

Qualifications

• Minimum 5+ years of programming in Python and/or C/C++.
• Experience developing software for real robots.
• Strong background in non-GPS localization (INS, EKF/UKF, SLAM, odometry).
• Deep understanding of localization algorithms beyond typical packages.
• Proven track record from implementation to measured improvement.
• CS degree or equivalent coding excellence.
• Excellent communication skills.
• Interest in multi-planet human exploration.

Responsibilities

• Own the localization architecture from approach to implementation and testing.
• Develop software estimating rover position from sensors (IMU, cameras, LIDAR, wheel odometry, star trackers, sun sensors, terrain matching).
• Handle calibration, time synchronization, coordinate frames, and sensor health across the sensor suite.
• Test on rover prototypes at a lunar analog test site and in Gazebo-based mission simulations built on NASA lunar terrain data.
• Define metrics and ground truth to quantify drift, accuracy, and consistency, and use field results to drive technical decisions.
• Integrate localization outputs and health signals with path planning, hazard avoidance, and recovery behavior.

Skills

Python

C/C++

Robotics software development

Non-GPS localization experience

System integration and testing

Strong communication

Education

Bachelor's degree in Computer Science or related field

Tools

ROS/ROS2

Gazebo

MuJoCo

Job description

About the Company

Our client is a venture-backed space technology company building the infrastructure that will make permanent human settlements on the Moon and Mars possible. It is already generating significant commercial revenue from its hardware products and is using that momentum to fund its lunar robotics program: autonomous rovers designed to produce essential resources on the Moon's surface.

The Role

You'll own the software that determines and tracks where the company's lunar rovers are as they drive across the Moon. There is no GPS, prior map data is low resolution, lighting swings between blinding sun and deep shadow, the terrain offers few vertical features, and wheels slip in loose regolith. Solving localization under these conditions takes both a deep understanding of how estimation and mapping algorithms actually work and hands-on experience making them hold up on real robots outside the lab.

The right person can choose a technical approach, write the software, test it on hardware and in simulation, measure where it performs well and where it breaks, decide whether to refine or pivot, and run that entire cycle quickly.

What You'll Do

• Own the localization architecture for the rover, from technical approach through implementation, testing, and iteration
• Develop software that estimates rover position from sensors such as IMU, cameras, LIDAR, wheel odometry, star trackers, sun sensors, and terrain matching
• Handle calibration, time synchronization, coordinate frames, and sensor health across the sensor suite
• Test on rover prototypes at a lunar analog test site and in Gazebo-based mission simulations built on NASA lunar terrain data
• Define metrics and ground truth to quantify drift, accuracy, and consistency, and use field results to drive technical decisions
• Integrate localization outputs and health signals with path planning, hazard avoidance, and recovery behavior

What You Bring

• 5+ years of programming experience in Python and/or C/C++
• Experience developing software that runs on real robots
• Substantive experience with non-GPS localization (e.g., INS, EKF/UKF, factor graphs, SLAM, odometry)
• A deep understanding of how localization algorithms work, beyond configuring existing packages
• A track record of taking a system through implementation, testing, debugging, and measured improvement
• A degree in Computer Science or a related field, or demonstrated coding excellence through significant professional experience
• Outstanding communication skills
• A genuine desire to help make humanity a multi-planet species

Nice to Have

• Localization on outdoor robots or autonomous vehicles, especially in natural terrain
• LIDAR-based localization (LIDAR odometry, LIDAR SLAM, scan matching)
• Vision-based localization (visual odometry, VIO, VSLAM)
• Terrain or map matching using elevation or imagery data
• ROS/ROS2 and simulators such as Gazebo or MuJoCo
• Aerospace software or production deployment experience

You Don't Need Space Experience

Some of the best candidates for this role will come from other demanding environments: off-road, agricultural, mining, or defense robots; GNSS-denied autonomous vehicles; drones and visual-inertial navigation; underwater or subterranean robots; LIDAR mobile mapping; or research teams that deployed their algorithms in the field. What matters is real-robot ownership and strong fundamentals.

How the Team Works

• Extreme ownership: You own problems from root cause to delivered solution.
• Bias toward action: When something blocks progress, you find a way around it instead of waiting.
• Mission over ego: Engineering requires disagreement, and the best idea wins regardless of who it came from.

Eligibility

This position requires access to export-controlled information. Candidates must be a U.S. citizen or national, U.S. lawful permanent resident, refugee under 8 U.S.C. § 1157, asylee under 8 U.S.C. § 1158, or otherwise eligible to obtain the required authorizations from the U.S. Department of State.