NVIDIA
Director, Perception - Autonomous Vehicles
$320K–$489K
US, CA, Santa Clara · Director · Full-time · Posted Sep 10, 2026
- Experience
- 10+ overall years, including at least 5+ years in a senior leadership role.
- Education
- Ph.D. or master's degree in computer science, Robotics, Artificial Intelligence, visual computing, or a related field (or equivalent experience).
About the job
Leads the organization and technical development of production autonomous-vehicle perception systems, including deep-learning models for sensor fusion and real-time 3D world reconstruction.
What you'll do
- Manage, mentor, and grow teams of engineering managers, researchers, and software engineers.
- Set architecture direction and oversee the machine-learning lifecycle through deployment on NVIDIA DRIVE platforms.
- Coordinate with Planning and Control, Mapping, Hardware, and Safety teams; ensure integration and automotive quality and safety standards.
What they ask for
- Ph.D. or master's degree in a related field, or equivalent experience.
- 10+ years of industry experience in deep learning, computer vision, or autonomous robotics, including 5+ years in senior leadership managing large teams or multiple management layers.
- Hands-on expertise in modern network architectures and multi-sensor fusion.
- Track record deploying production-grade, safety-critical machine-learning models to edge or embedded platforms; foundational knowledge of Python, C++, and deep-learning frameworks.
Nice to have
- Success shipping autonomous-vehicle software to mass production; top-tier publications or significant open-source contributions.
- Experience with end-to-end autonomous driving, Large Vision Models, sim-to-real transfer, or hardware-software co-design for NVIDIA GPUs.
Pay
320,000 USD - 488,750 USD (base salary)
Tools
Transformers, BEV, Occupancy Networks, Vision-Language-Action models, CNNs, Python, C++, PyTorch
AI focus
Leads development of deep-learning perception models, multi-sensor fusion, 3D world reconstruction, and the machine-learning lifecycle from data generation and training through acceleration and deployment.
Model-written summary · the original posting is authoritative