Senior Perception Software Engineer

NVIDIA•US, CA, Remote, WA

1d•$224,000 - $356,500

About The Position

NVIDIA is seeking a Software Engineer to design, develop, and productize autonomous driving solutions as part of its perception team. The role involves building perception software components and improving their robustness, efficiency, and reusability in collaboration with perception engineers and researchers. The engineer will deploy state-of-the-art perception deep neural network models to embedded platforms and analyze system performance using real and synthetic benchmark data.

Requirements

BS, MS, or PhD in Computer Science or a related field, or equivalent experience.
12+ years of proven work experience on embedded platforms for real-time applications as a Software Engineer or Software Developer.
Strong programming skills in C++ and Python.
Experience with test-driven development.
Experience with a disciplined, design-based approach to software development.
Proficiency in software engineering tools.
Ability to document requirements and specifications.
Strong communication and collaboration skills.

Nice To Haves

Experience with CUDA or SIMD.
Hands-on experience deploying DNN models to embedded platforms for real-time applications.
Hands-on experience using simulated or synthetic data to improve machine learning model accuracy.
Background with deep learning frameworks such as TensorFlow or PyTorch.
Experience with integration and testing on real hardware or robotic platforms.

Responsibilities

Execute the full software development life cycle (SDLC).
Write well-designed, testable code.
Integrate perception and other software components into a fully functional software system.
Develop software verification plans and quality assurance procedures.
Document and maintain software functionality.
Solve problems, debug, and upgrade existing systems.
Stay aligned with project plans and industry standards.
Deploy state-of-the-art perception DNN models to embedded platforms.
Analyze and improve the system using real and synthetic benchmark data.