What you’ll be doing:

• Collaborate with cross-functional teams to understand requirements and contribute to the design and architecture of NVIDIA DirectML drivers.
• Develop and implement high-quality, efficient, and reliable NVIDIA DirectML drivers following the best software development practices adhering to industry standards.
• Identify and resolve performance bottlenecks for cutting-edge deep learning models in computer vision, generative AI, and other related fields.
• Debug and resolve driver-related issues to ensure smooth functionality across multiple GPU architectures and platforms.
• Collaborate with Microsoft to drive the advancements in ONNX and DML runtimes and ensure seamless integration with NVIDIA DirectML driver.
• Build and contribute to performance profiling tools that improves the efficiency of the performance analysis workflow.
• Ensure the effective deployment of essential directed tests and models through collaboration with the automation team, thereby ensuring the robustness of automated testing.

What we need to see:

• Bachelor's, Master's, or PhD in Electrical Engineering, Computer Science/Engineering, Mathematics, Physics, or a related field (or equivalent experience).
• 5+ years of industry experience with proficiency in developing and optimizing DirectML drivers or similar graphics and machine learning drivers such as DirectX, Vulkan, or CUDA.
• Familiarity with GPU hardware architecture, shader programming, and parallel processing techniques.
• Excellent C++ programming and debugging skills with a strong understanding of data structures and algorithms.
• Familiarity with deep learning frameworks such as PyTorch and TensorFlow and exposure to large language models and transformer architectures.
• Strong analytical and problem-solving abilities, with the capacity to multitask effectively in a multifaceted environment.
• Outstanding written and oral communication skills, enabling effective collaboration with management and engineering teams.