SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Aurora is building self-driving technology to transform mobility and logistics. The Map Fusion team is seeking a Staff Software Engineer to architect and lead the development of real-time algorithms that enable autonomous vehicles to detect and respond to dynamic environmental changes—such as construction zones, emergency vehicles, and traffic control changes—with safety-critical precision.
In this role, you will own the technical roadmap for onboard map-update algorithms that process sensor data in real-time under strict embedded performance and safety constraints. You'll define the architecture for low-latency compute solutions and partner closely with leadership and principal engineers across Motion Planning, Localization, Perception, and Fleet Intelligence to ensure seamless end-to-end integration.
Key responsibilities include architecting graph-centric and geometric algorithms for accurate lane-line and drivable-surface geometry, spearheading safety-case formulation for machine learning and classical fusion methods, and providing technical mentorship and rigorous design reviews to foster engineering excellence within the team.
You will need extensive professional experience architecting and optimizing high-performance, production-grade C++ in resource-constrained environments, strong Python proficiency for rapid prototyping and algorithm testing, and deep expertise in 3D geometric reasoning, coordinate transforms, and spatial data at scale. Mastery of graph theory, network optimization, and complex search algorithms is essential. You should have a demonstrated track record of leading complex technical projects from concept to production, aligning multiple stakeholders, and mentoring engineers.
Desirable qualifications include deep expertise in deep learning architectures for spatial/temporal/sensor data and deploying them safely on real-world systems, as well as mastery of Bayesian inference and Kalman filtering techniques.
