System Software Engineer - Deep Learning

NVIDIA seeks a System Software Engineer - Deep Learning for its Tegra Solutions Engineering Business Unit in Bengaluru, India. The role supports the NVIDIA DRIVE AI platform for autonomous driving, in-cabin functions, driver monitoring and other automotive safety features, with work spanning autonomous driving and in-car infotainment.

Responsibilities

• Develop solutions using NVIDIA GPUs and deep learning accelerators (DLAs) to accelerate inference for advanced driver-assistance systems (ADAS).
• Build SDKs and frameworks to accelerate LLMs and other state-of-the-art models on the NVIDIA DRIVE platform.
• Benchmark and evaluate models to improve inference latency, accuracy and power consumption. Follow deep learning research and experiment with improvements to NVIDIA’s automotive deep neural networks.
• Manage technical relationships with automotive customers, help them develop solutions using NVIDIA technology, and collaborate with engineering teams across the US, APAC, India and Europe.

Requirements

• BS or MS in Computer Science, Computer Engineering or Electrical Engineering.
• At least 5 years of experience developing or using deep learning frameworks such as TensorFlow, Keras, PyTorch, Caffe or ONNX.
• Proven experience optimizing DNN layers for GPUs or other DSPs; understanding of compiler infrastructure such as LLVM and MLIR and optimization flows for deep learning accelerators.
• Proficiency in C, C++ and data structures; strong operating-system fundamentals and knowledge of CPU/GPU architecture; familiarity with CNN, LSTM and Transformer architectures.

The posting lists the position as full-time and mentions competitive salaries and a comprehensive benefits package, without specifying a salary amount.