Your responsibilities will include:

• Training, fine-tuning, optimizing, and customizing perception DNNs in low precision (FP16/INT8)
• Applying sophisticated quantization of DNNs
• Improving DNN architectures using ML algorithms on NVIDIA GPUs or DLAs
• Continuously improving inference speed, accuracy, and power consumption of DNNs
• Staying up to date with the latest research and innovations in deep learning, implementing, and experimenting with new insights to improve NVIDIA's automotive DNNs.

To succeed in this role, you will need:

• A Master's or Ph.D. degree in Computer Science, Computer Vision, Computer Architecture, or equivalent experience in a technical field
• 5+ years of work experience in software development
• 2+ years of experience in developing or using deep learning frameworks (e.g., PyTorch, JAX, TensorFlow, ONNX, etc.)
• Experience with solving a computer vision task using deep neural networks, such as object detection, scene parsing, image segmentation
• Strong Python and/or C/C++ programming skills
• Proven technical foundation in CPU and GPU architectures, containers (nvidia-docker), numeric libraries, modular software design
• Familiarity with CNNs and Transformer architectures
• Willingness to take action and have strong analytical skills
• Strong time-management and organization skills for coordinating multiple initiatives, priorities, and implementations of new technology and products into very sophisticated projects.

If you have experience with low precision inference, quantization, compression of DNNs, or open-source project ownership or contribution, you will stand out from the crowd.

XML job scraping automation by YubHub