Principal Software Engineer, Mapping - Autonomous Vehicles

NVIDIA seeks a Principal Software Engineer to build planet-scale maps supporting self-driving technology using crowdsourced data from millions of cars. The role involves computer vision, geometry, pose estimation, sensor fusion, and large-scale systems.

The engineer will transform sparse perception signals and dense video clips into accurate and fresh maps. These maps help autonomous vehicles understand the road with high-quality map priors, improving driving performance, safety, and coverage.

Responsibilities:

• Build scalable mapping systems using crowdsourced perception data and multi-camera video collected from a vast number of vehicles.
• Develop 3D reconstruction, structure-from-motion, pose estimation, and multi-view geometry algorithms for large-scale road scene understanding.
• Build map fusion and change-detection methods that handle noisy observations, dynamic scenes, imperfect localization, and global consistency constraints.
• Develop C++ production systems and offline pipelines that transform fleet data into reliable map products used in self-driving and driver support technologies.
• Invent evaluation methods to measure map accuracy, freshness, coverage, consistency, and downstream autonomy impact.
• Develop visualization, debugging, and triage tools to understand reconstruction quality, map issues, localization errors, and fleet data gaps.
• Work closely with perception, localization, simulation, planning, and infrastructure teams to integrate crowdsourced maps into autonomous driving systems.
• Improve the scale, fidelity, freshness, and reliability of maps built from real-world fleet data.

Requirements:

• Strong programming skills in C++ and experience building production-quality software systems.
• Solid foundation in 3D computer vision, 3D geometry, multi-view geometry, structure from motion, SLAM, pose estimation, or related areas.
• Experience working with large-scale sensor data, including camera video, perception outputs, vehicle poses, GPS/IMU signals, lidar, radar, or map data.
• Ability to reason about coordinate frames, calibration, uncertainty, optimization, geometric consistency, and error propagation.
• Experience crafting algorithms that are robust to noisy real-world data, dynamic objects, occlusions, incomplete coverage, and long-tail failures.
• Strong debugging and analytical skills, including the ability to inspect data visually, build metrics, and connect system-level failures to algorithmic root causes.
• 15+ years of experience.
• BS, MS, or PhD in Computer Science, Robotics, Electrical Engineering, Mathematics, or a related technical field (or equivalent experience).

Benefits:

• Highly competitive salaries.
• Comprehensive benefits package.
• Equity eligibility.

The application deadline for this job is June 30, 2026.