NVIDIA is seeking exceptional Senior AI Training Performance Engineers to optimize AI training workloads, one of the most critical applications in modern computing. This role combines deep technical expertise in GPU architecture, deep learning frameworks, and performance optimization to achieve maximum efficiency in AI training systems. The position offers a unique opportunity to impact both hardware and software development at NVIDIA, a leading force in the AI revolution.
The role involves working across the entire hardware/software stack, from GPU architecture to Deep Learning Frameworks, to achieve peak performance. You'll be responsible for analyzing and optimizing state-of-the-art neural networks, implementing production-quality software in NVIDIA's deep learning platform stack, and working with proprietary processor and system simulators for future architecture studies.
As part of NVIDIA's Deep Learning Architecture team, you'll be at the forefront of AI computing, working with cutting-edge technology and contributing to products that are transforming industries worldwide. The company is widely recognized as one of the technology world's most desirable employers, known for its innovative work in AI computing and accelerated computing solutions.
The ideal candidate should possess a strong academic background (PhD or MS with equivalent experience) in Computer Science or related fields, combined with extensive experience in deep learning, computer architecture, and performance optimization. This role requires expertise in programming languages such as C++, Python, and CUDA, along with a deep understanding of GPU architecture and neural network training processes.
This position offers the opportunity to work on industry-leading Deep Learning products while helping users worldwide achieve faster training speeds. You'll be joining a forward-thinking team in a company that's leading the AI revolution, making this an excellent opportunity for those passionate about performance optimization and artificial intelligence.