We are seeking a Senior Software Engineer to help define the runtime intelligence and safety architecture behind next-generation autonomous driving systems. This role sits at the intersection of end-to-end AI driving models, vehicle dynamics, and safety-critical autonomy. Modern AI models can generate highly capable driving behaviors, but deploying them safely in production vehicles requires solving some of the hardest problems in real-time robotics: compute constraints, physical feasibility, uncertainty handling, runtime validation, and safety arbitration. You will build the framework that bridges large-scale learned driving models with deterministic planning and vehicle-level safety guardrails-ensuring AI-generated trajectories are physically feasible, safe, explainable, and deployable on real automotive hardware platforms. This role is ideal for engineers excited about bringing modern AI into real-world physical systems where latency, compute efficiency, vehicle dynamics, and safety constraints fundamentally matter. What You'll Be Doing: • Design and integrate planning frameworks that combine end-to-end learned driving models with classical trajectory planning and deterministic safety systems. • Develop runtime arbitration and safety enforcement mechanisms between AI-generated trajectories and rule-based safety constraints. • Build scalable architecture enabling large AI driving models to operate reliably within automotive compute, latency, and real-time execution constraints. • Develop execution frameworks that ensure AI-generated behaviors satisfy vehicle dynamics, collision avoidance, passenger comfort, and safety requirements in real time. • Define and implement safety-oriented planning capabilities including trajectory validation, fallback handling, runtime policy gating, and Minimum Risk Maneuver (MRM) strategies. • Partner closely with AI, planning, controls, and systems teams to productize learned driving models into deployable autonomous vehicle systems. • Analyze and debug complex autonomy edge cases involving uncertainty, model failure modes, planner disagreement, and real-world safety constraints. • Improve observability, reliability, and debuggability across large-scale autonomy planning systems operating in simulation and on-vehicle environments. • Drive architectural decisions balancing AI capability, system robustness, safety, and embedded deployment efficiency. • Influence next-generation autonomy architecture defining how foundation-model and learning-based driving systems coexist with production-grade safety-critical vehicle platforms. What We Need To See: • BS, MS, or PhD (or equivalent experience) in Computer Science, Robotics, Electrical Engineering, AI/ML, or related technical field. • 12+ years of relevant industry experience in autonomous systems, robotics, AI infrastructure, or safety-critical software systems. • Strong software engineering fundamentals with production C++ development experience. • Strong understanding of autonomous vehicle planning, trajectory generation, motion planning, or robotics systems. • Experience working with machine learning systems and understanding how learned models behave under uncertainty and real-world edge cases. • Experience delivering scalable, production-quality systems from architecture through deployment. • Strong debugging, systems integration, and performance optimization skills for real-time systems. • Excellent communication and cross-functional technical leadership abilities. Ways To Stand Out From The Crowd: • Experience deploying machine learning models into real-time embedded or robotics systems. Deep understanding of both classical planning systems and end-to-end learning approaches for autonomous driving. • Experience with runtime safety validation, fallback systems, policy gating, or safety arbitration frameworks. • Familiarity with foundation-model-based driving systems, learned planners, generative trajectory models, or AI-native autonomy stacks. • Strong intuition for bridging the gap between offline AI model capability and production deployment constraints. Experience with large-scale autonomy simulation, scenario replay, evaluation infrastructure, or safety validation pipelines. • Passion for solving deeply challenging engineering problems at the intersection of AI, robotics, and real-world deployment. Your base salary will be determined based on your location, experience, and the pay of employees in similar positions. The base salary range is 224,000 USD - 356,500 USD. You will also be eligible for equity and benefits. Applications for this job will be accepted at least until May 22, 2026. This posting is for an existing vacancy. NVIDIA uses AI tools in its recruiting processes.