Senior Radar Perception Engineer, Obstacle Foundation Models – Autonomous Vehicles

Details

NVIDIA is seeking an exceptional Senior Radar Perception Engineer to help design and productize its next-generation autonomous driving perception stack. You will work on the core 3D radar and multimodal obstacle perception pipeline, contribute to architecture and algorithm design, and remain deeply hands-on with implementation, including transformer-based radar foundation models and multi-sensor fusion techniques.

Responsibilities

• Develop and improve the technical design, architecture, and roadmap for radar-based 3D obstacle perception supporting end-to-end autonomous driving functionalities.
• Conduct applied research on deep learning models for radar point cloud data, addressing low and non-uniform angular resolution, multipath and ghost targets, micro-Doppler signatures, and severe class imbalance.
• Explore weakly supervised pretraining and improve radar perception using large auto-labeled datasets.
• Design and implement advanced 3D perception models using radar inputs, including range-Doppler and azimuth-elevation maps as well as sparse and dense point clouds.
• Develop multi-sensor fusion solutions using camera, radar, and lidar for obstacle detection, tracking, and Bird’s-Eye-View scene understanding.
• Drive radar sensor evaluation, selection, and layout optimization for L2–L4 autonomous driving applications.
• Build efficient, production-grade deep learning models, including large-scale radar pretraining, cross-modal distillation such as lidar-to-radar, and parameter-efficient fine-tuning such as LoRA.
• Define and maintain KPI frameworks, analyze real and synthetic datasets, identify radar-specific failure modes, and improve accuracy, robustness, and efficiency.
• Contribute to radar data strategy, labeling requirements, data collection, annotation prioritization, active learning, automated radar labeling, and model-in-the-loop tooling.
• Collaborate with safety, systems, software, data, labeling, and ground-truth teams to productize solutions that meet requirements for safety, latency, resource usage, and software robustness.

Requirements

• 12+ years of hands-on experience developing deep learning-based perception, radar signal processing, or closely related systems for complex real-world problems.
• Strong proficiency with deep learning frameworks such as PyTorch and experience taking models from prototype to production.
• Experience with data-driven development and collaboration with data, labeling, and ground-truth teams.
• Strong programming skills in Python and/or C++, including reliable, high-performance, production-quality software development.
• Excellent communication and collaboration skills across multidisciplinary AI, hardware, and safety engineering teams.
• BS, MS, or PhD in Computer Science, Electrical Engineering, Robotics, or a related field, or equivalent experience.

Preferred Qualifications

• Experience designing and deploying radar-based or multimodal perception solutions for autonomous driving or robotics at scale.
• Experience deploying DNN-based perception pipelines on embedded or real-time platforms and optimizing latency, memory, and compute usage.
• Familiarity with Transformers and BEV networks.
• Deep understanding of radar physics and digital signal processing, including FMCW, beamforming, CFAR, and micro-Doppler.
• Strong publication record or recognized contributions in deep learning, radar perception, multisensor fusion, or autonomous systems.
• Experience with CUDA development, custom CUDA kernels, and GPU-accelerated training or inference pipelines.

Compensation and Benefits

The base salary range is $224,000–$356,500 USD, determined by location, experience, and compensation of employees in similar positions. The role also includes eligibility for equity and benefits.