Senior Deep Learning Compiler Engineer - PyTorch

NVIDIA is seeking a Senior Deep Learning Compiler Engineer to join their innovative team working on Thunder, a cutting-edge source-to-source compiler designed to optimize PyTorch models for NVIDIA GPUs. This role sits at the intersection of AI, compiler technology, and high-performance computing, offering a unique opportunity to shape the future of accelerated computing.

The position involves working directly with the engineers who built PyTorch for NVIDIA hardware, focusing on developing compiler technologies that accelerate massive deep learning workloads. You'll be part of a collaborative environment, working alongside experts from various compiler and systems teams, including those behind nvFuser, TVM, XLA, and CUDA.

The ideal candidate brings 8+ years of experience, strong Python skills, and deep knowledge of compiler concepts and deep learning frameworks. This role offers the chance to make significant contributions to open-source projects that will impact the entire AI ecosystem. You'll be working with modern compiler stacks like PyTorch 2.0's TorchDynamo and TorchInductor, creating solutions that enhance performance and accessibility for AI developers worldwide.

At NVIDIA, you'll be part of a company at the forefront of breakthroughs in Artificial Intelligence and High-Performance Computing, offering competitive compensation, comprehensive benefits, and a diverse, inclusive work environment. This role provides an exceptional opportunity to work on cutting-edge technology while contributing to solutions that benefit the broader AI community.