GPU Compiler LLVM Backend Intern - 2027

NVIDIA's invention of the GPU 1999 sparked the growth of the PC gaming market, redefined modern computer graphics, and revolutionized parallel computing. More recently, GPU deep learning ignited modern AI — the next era of computing — with the GPU acting as the brain of computers, robots, and self-driving cars that can perceive and understand the world. Today, we are increasingly known as “the AI computing company”.

We are searching for a LLVM Compiler Intern for an exciting and fun role in our GPU Software organization. Our Compiler team is responsible for constructing and emitting the highest performance GPU machine instructions for Graphics (OpenGL, Vulkan, DX) and Compute (CUDA, PTX, OpenCL, Fortran, C++). This team is comprised of worldwide leading compiler engineering experts who provide leading edge performance and capabilities for NVIDIA's current and future complex parallel SIMT architectures.

What You Will Be Doing

• Understand, modify, and improve an NVIDIA GPU compiler and assembler written in C++
• Design and develop new compiler passes and optimizations to produce best-in-class, robust, supportable compiler and tools
• Work with global compiler, hardware and application teams to oversee improvements and problem resolutions
• Be part of a team that is at the center of deep-learning compiler technology spanning architecture design and support through functional languages

What We Need To See

• Pursuing B.S. or higher degree in Computer Science/Engineering
• Excellent hands-on C++ programming skills
• Strong background in software engineering principles with a focus on crafting robust and maintainable solutions to challenging problems
• Good communication and documentation skills and self-motivated

Ways To Stand Out From The Crowd

• Masters or PhD preferred, with some compiler code generation experience
• Experience in developing CUDA, DirectX, OpenGL/Vulkan applications
• Experience in performance analysis and compile time analysis tools
• Exposure to various ML techniques