Senior Software Engineer - Autonomous Vehicles

NVIDIA•Santa Clara, CA

About The Position

Build and own core online mapping software that operates in autonomous vehicles, from architecture and implementation through deployment. The system transforms raw sensor data into an accurate, live 3D model of lanes, boundaries, crosswalks, intersection structures, and traffic-control semantics for the driving stack. The role focuses on turning mapping algorithms into safe, fast, maintainable production software rather than developing algorithms from scratch. Software will run on NVIDIA DRIVE Orin and Thor, meet ASIL-B/D requirements, and ship across L2++, L3, and L4 programs. The engineer will work closely with algorithm engineers and collaborate across the AV stack to understand, implement, optimize, and test algorithmic solutions. The role emphasizes production C++ for real-time embedded systems, correctness, determinism, maintainability, reusable architecture, and performance optimization of C++ and CUDA code. It also requires safety-focused development, static analysis, thorough unit and integration testing, and robust handling of real-time vehicle signals under varied conditions and failure modes. The target is 100% line and branch coverage. The expected base salary is $152,000–$241,500 for Level 3 and $184,000–$287,500 for Level 4. The employee will also be eligible for equity and benefits. Applications will be accepted at least until November 9, 2026. This posting is for an existing vacancy.

Requirements

Bachelor’s or master’s degree in Computer Science, Electrical Engineering, or a related field, or equivalent experience.
At least 5 years of validated experience building robust, production-quality software.
At least 3 years of professional C++ development experience, including modern C++, memory management, concurrency, and performance.
Experience designing software architecture and APIs for large, long-lived codebases.
Quality-first approach and comfort with static analysis, unit testing, and high-coverage test strategies.
Solid understanding of algorithms and data structures, and ability to work with algorithm developers and understand their work.
Strong communication and analytical skills, with the ability to work effectively across multiple teams.
Ability to work autonomously and responsibly, demonstrate outstanding teamwork, and enjoy solving difficult problems.

Nice To Haves

Software development experience on embedded, real-time, or automotive platforms, including real-time operating systems such as QNX or Linux RT.
Experience with functional safety standards and processes, including ISO 26262 and ASIL-B/D, MISRA or AUTOSAR C++ coding guidelines, and safety-qualified toolchains.
Experience achieving high or full structural coverage on production code, including line and branch coverage and ideally MC/DC.
Experience with GPGPU programming using CUDA and GPU performance tuning.
Strong runtime profiling and CPU and GPU code optimization skills.

Responsibilities

Build and own core online mapping software operating in the car, from architecture through implementation to deployment.
Write production C++ for real-time embedded systems, focusing on correctness, determinism, and maintainability.
Build software architecture and frameworks that promote code reuse, clean interfaces, and long-term maintainability across vehicle programs.
Improve the performance of C++ and CUDA code on NVIDIA DRIVE Orin and Thor using profilers to identify and address bottlenecks and meet real-time budgets.
Use CUDA GPU parallelism to make compute-heavy mapping components fast and efficient.
Implement and incorporate algorithmic solutions from the mapping team into NVIDIA’s AV stack, collaborating with algorithm engineers to understand and improve them.
Develop according to automotive safety standards, including ASIL-B/D, use thorough static analysis, and adhere to safety-focused coding guidelines.
Develop detailed unit and integration tests to achieve 100% line and branch coverage.
Handle real-time system signals from vehicles operating around the world and build software robust to varied conditions and failure modes.
Collaborate with perception, localization, planning, and infrastructure groups to incorporate online mapping into the AV stack across L2++, L3, and L4 programs.