Senior Perception Architect - Autonomous Driving (Equity)

NVIDIA Corporation; Santa Clara (CA); Hybrid; USD 272,000 - 431,000; Full time

Benefits offered by this job

Equity

Benefits

Job summary

NVIDIA Corporation is seeking a Principal Perception Engineer to lead the design and productization of the next-generation autonomous driving perception stack. This senior IC will own the technical direction for 3D obstacle perception and drive cross-functional execution, mentoring engineers while staying deeply hands-on with architecture, algorithms, and implementation.

You will define and drive production-grade models, leverage multi-sensor fusion and modern transformer-based 3D perception,

Qualifications

• 15+ years in deep learning–based perception or related systems for real-world problems.
• Demonstrated technical leadership as senior/principal contributor: own features/subsystems end-to-end.
• Data-driven development with collaboration with data/ground-truth teams.

Responsibilities

• Own the technical vision, architecture, and roadmap for 3D obstacle perception for autonomous driving.
• Design and develop 3D perception models using multi-camera and multi-sensor inputs.
• Lead production-grade DL models, define objectives, architectures, training, and evaluation best practices.
• Define KPI frameworks, analyze datasets, and improve accuracy, robustness, and efficiency.
• Lead data strategy: data collection, labeling, and model-in-the-loop tooling; active learning and auto-labeling.

Skills

Deep learning

3D perception

PyTorch

Python

C++

Technical leadership

Communication

Education

BS/MS/PhD in CS/EE or related

Tools

CUDA

Linux

Job description

NVIDIA Corporation is seeking a Principal Perception Engineer to lead the design and productization of the next-generation autonomous driving perception stack. This senior IC will own the technical direction for 3D obstacle perception and drive cross-functional execution, mentoring engineers while staying deeply hands-on with architecture, algorithms, and implementation.

You will define and drive production-grade models, leverage multi-sensor fusion and modern transformer-based 3D perception,