Principal Mapping & Localization Engineer

Fruition Group US; San Francisco (CA); On-site; USD 200,000 - 260,000; Full time

Benefits offered by this job

Equity

Job summary

Fruition Group US is partnering with an autonomous trucking technology company to hire a Principal Engineer, Mapping and Localization in the San Francisco Bay Area. You will shape architecture, decide build-vs-buy, and set the long-term roadmap for localization and mapping, while remaining hands-on with the hardest problems in the stack.

You will lead sensor fusion and odometry design, optimize for real-time operation, and mentor junior engineers as you scale map data across a fleet.

Qualifications

• Master's or PhD in Robotics, Computer Science, Electrical Engineering, or a closely related technical field.
• 7+ years in the robotics industry, including at least 4 years specifically building production localization or mapping systems.
• Deep, expert-level fluency in modern C++ (C++14/17/20), with strong command of memory management in real-time systems.
• Solid grounding in the math of probabilistic robotics - comfortable with Kalman filtering, nonlinear optimization tools like Ceres, g2o, or GTSAM, and SLAM techniques.
• Comfortable working with 3D geometry, linear algebra, and coordinate frame transforms such as quaternions and SE(3).
• Hands-on experience with common robotics libraries such as Eigen, Ceres, PCL, OpenCV, or GTSAM.
• A plus if you've deployed localization systems on autonomous vehicles or mobile robots, worked with approaches like LOAM, LeGO-LOAM, ORB-SLAM, or VINS, or have experience optimizing for CUDA/GPU or ARM-based embedded platforms.

Responsibilities

• Set the overall technical direction for localization and mapping, and decide where to build in-house versus adopt existing tools.
• Architect sensor fusion and odometry solutions suited to real-time operation, drawing on techniques such as factor graphs, Kalman-filter variants, and particle filters.
• Build out the systems responsible for creating, updating, and serving detailed map data across the fleet.
• Develop methods for localizing against known reference data and matching what the vehicle sees to features already recorded on the map.
• Keep maps from going stale by building systems that use data gathered across the fleet to flag real-world changes - new construction, updated signage, and the like.
• Tune algorithm performance for the compute hardware actually running onboard, weighing trade-offs between speed, memory footprint, and power draw.
• Give the localization system the ability to recognize and flag when its own confidence is dropping.
• Help grow more junior engineers through mentorship and hands-on code review.

Skills

C++ (modern)

Kalman filtering

SLAM techniques

Probabilistic robotics

3D geometry

Real-time systems

Education

Master's or PhD in Robotics, CS, or EE

Tools

Eigen

Ceres

PCL

OpenCV

GTSAM

CUDA

Job description

Principal Engineer, Mapping and Localization

Location: San Francisco Bay Area, CA (on-site) $200,000 - $260,000 + Equity

About the Role

FruitionGroup is partnering with an autonomous trucking technology company to find their next Principal Engineer, Mapping and Localization. Our client needs their self-driving vehicle to always know precisely where they sit on the road, down to a few centimeters, by blending data from multiple sensor types - lidar, radar, cameras, inertial sensors, and satellite positioning - together with detailed maps of the road network. The hard part is making that work reliably in places where GPS signal drops out, and on roads that are constantly being re-paved, re-signed, or torn up for construction. You'll shape the architecture, make the build-vs-buy calls, and set the long-term roadmap for localization and mapping, while staying hands-on with the hardest problems in the stack.

What You'll Do

• Set the overall technical direction for how the company approaches localization and mapping, and decide where to build in-house versus adopt existing tools.
• Architect sensor fusion and odometry solutions suited to real-time operation, drawing on techniques such as factor graphs, Kalman-filter variants, and particle filters.
• Build out the systems responsible for creating, updating, and serving detailed map data across the fleet.
• Develop methods for localizing against known reference data and matching what the vehicle sees to features already recorded on the map.
• Keep maps from going stale by building systems that use data gathered across the fleet to flag real-world changes - new construction, updated signage, and the like.
• Tune algorithm performance for the compute hardware actually running onboard, weighing trade-offs between speed, memory footprint, and power draw.
• Give the localization system the ability to recognize and flag when its own confidence is dropping.
• Help grow more junior engineers through mentorship and hands-on code review.

What Our Client Is Looking For

• Master's or PhD in Robotics, Computer Science, Electrical Engineering, or a closely related technical field.
• 7+ years in the robotics industry, including at least 4 years specifically building production localization or mapping systems.
• Deep, expert-level fluency in modern C++ (C++14/17/20), with strong command of memory management in real-time systems.
• Solid grounding in the math of probabilistic robotics - comfortable with Kalman filtering, nonlinear optimization tools like Ceres, g2o, or GTSAM, and SLAM techniques.
• Comfortable working with 3D geometry, linear algebra, and coordinate frame transforms such as quaternions and SE(3).
• Hands-on experience with common robotics libraries such as Eigen, Ceres, PCL, OpenCV, or GTSAM.
• A plus if you've deployed localization systems on autonomous vehicles or mobile robots, worked with approaches like LOAM, LeGO-LOAM, ORB-SLAM, or VINS, or have experience optimizing for CUDA/GPU or ARM-based embedded platforms.