Radar Perception Architect – Obstacle Foundation Models

NVIDIA Gruppe; Santa Clara (CA); On-site; USD 224,000 - 356,500; Full time

Job summary

NVIDIA is seeking an expert in radar perception and deep learning to lead architecture, research, and production deployment for autonomous driving radar systems in Santa Clara, CA. You will advance multi-sensor fusion with camera and lidar data, optimize models for low latency, and collaborate across AI, hardware, and safety teams.

The role emphasizes experimentation with large-scale radar pretraining, model distillation, and end-to-end deployment, requiring 12+ years in perception or related

Qualifications

• 12+ years of hands-on experience developing deep learning–based perception, radar signal processing, or closely related systems for complex real-world problems.
• Data-driven development with close collaboration with data, labeling, and ground-truth teams on radar data strategy and model improvement.
• Strong programming skills in Python and/or C++, with production-quality software.
• Excellent communication and collaboration across AI, hardware, and safety engineering teams.
• BS/MS/PhD in Computer Science, Electrical Engineering, Robotics, or related fields (or equivalent).

Responsibilities

• Architect and roadmap radar-based 3D obstacle perception for autonomous driving functionalities, using state-of-the-art DNN and transformer architectures.
• Conduct applied research to maximize radar point cloud information content across representations.
• Design and implement advanced 3D perception models with radar inputs and multi-sensor fusion (camera, radar, lidar).
• Drive radar sensor evaluation, layout optimization, and seamless sensor fusion for L2-L4 applications.
• Build production-grade deep learning models and experiments, including large-scale radar pretraining and model distillation.
• Define KPI frameworks and analyze large-scale datasets to improve radar perception accuracy and robustness.
• Contribute to data strategy and labeling requirements, collaborating with data and ground-truth teams.
• Collaborate with safety, systems, and software teams to meet product requirements and deploy at scale.

Skills

Deep learning perception

Radar signal processing

Python/C++

Cross-functional collaboration

Education

BS/MS/PhD in CS/EE/Robotics

Tools

CUDA

Job description

NVIDIA is seeking an expert in radar perception and deep learning to lead architecture, research, and production deployment for autonomous driving radar systems in Santa Clara, CA. You will advance multi-sensor fusion with camera and lidar data, optimize models for low latency, and collaborate across AI, hardware, and safety teams.

The role emphasizes experimentation with large-scale radar pretraining, model distillation, and end-to-end deployment, requiring 12+ years in perception or related