Drimva

Wayve

Principle Engineer, 3D Reconstruction

$296K–$335K

Sunnyvale, California USA +1 · Hybrid · Full-time · Posted Oct 13, 2026

Location
Sunnyvale, California USA
London, United Kingdom
Education
MS/PhD in Robotics, Computer Science, State Estimation, Sensor Fusion, SLAM, 3D Reconstruction, or a related technical field; or equivalent practical experience (desired).
Visa sponsorship
Visa sponsorship where applicable.

About the job

Leads the technical direction and development of an offline 3D reconstruction system that turns vehicle sensor data into quality-assessed 3D geometry and ego-motion estimates for validation, data enrichment, and model development.

What you'll do

  • Set the architecture and roadmap, develop early prototypes, and guide the transition to a robust internal platform.
  • Define interfaces, output representations, quality standards, and validation strategies for reconstruction outputs.
  • Lead and mentor engineers while coordinating requirements and technical tradeoffs across specialist teams.

What they ask for

  • Deep expertise in SLAM, 3D reconstruction, mapping, state estimation, or a related robotics/perception field.
  • Hands-on experience with lidar-based reconstruction, point-cloud registration, pose estimation, and real-world robotics sensor data.
  • Experience building offline reconstruction, mapping, or perception systems for large volumes of sensor data and taking prototypes into reliable production systems.
  • Strong software engineering skills in C++, Python, or similar robotics/perception environments.

Nice to have

  • MS/PhD in a related technical field, or equivalent practical experience; autonomous vehicle, robotics, mapping, or ground-truth pipeline experience.
  • Experience with position estimation, nonlinear optimization, machine learning for 3D reconstruction, or robotics tooling such as ROS, PCL, OpenCV, or CUDA.

Pay

$295.5K – $335.3K (base salary)

Tools

C++, Python, GNSS/INS, ROS, PCL, OpenCV, CUDA, ICP

AI focus

Develops offline 3D reconstruction and ground-truth outputs from vehicle sensor data to support validation, data enrichment, and model development; the posting also notes desired experience applying machine learning to reconstruction, mapping, or scene understanding.

Model-written summary · the original posting is authoritative

First seen Oct 14, 2026 · Last verified Nov 5, 2026 (UTC)