Senior Radar Perception Engineer, Obstacle Foundation Models - Autonomous Vehicles
Core
Design and productize NVIDIA's next-generation autonomous driving perception stack, focusing on 3D radar and multi-modal obstacle perception pipelines using transformer-based foundation models and multi-sensor fusion.
Role type
Senior IC machine-learning engineer (radar perception & multi-modal fusion)
Builds
Production-grade deep learning models for obstacle detection, tracking, and Bird's-Eye-View (BEV) scene understanding
Domain
Autonomous vehicles, radar signal processing, deep learning
Deliverable
production ML models
Required skills
Radar signal processing, deep learning model design, multi-sensor fusion, transformer architectures, embedded optimization, CUDA development, data strategy, Python, C++, PyTorch
Preferred skills
Radar physics fundamentals (FMCW, beamforming, CFAR, micro-Doppler), large-scale pretraining, cross-modal distillation, parameter-efficient fine-tuning (LoRA), active learning, automated labeling, GPU acceleration
Technologies
PyTorch, CUDA, Transformers, BEV networks, FMCW, LoRA
Responsibilities
Develop technical architecture and roadmap for radar-based 3D obstacle perception; conduct applied research on deep learning models for radar point clouds; design and implement 3D perception models with multi-sensor fusion; drive radar sensor evaluation and layout optimization; build efficient production-grade deep learning models; define and maintain KPI frameworks for performance analysis; contribute to data strategy and auto-labeling workflows; collaborate with safety and systems teams for productization
Seniority
Senior, hands-on IC
