NVIDIA is looking to improve the performance of their GPU compiler backend to support current and future complex parallel SIMT architectures and drive the growth of the AI computing market
Requirements
- Excellent hands-on C++ programming skills
- Strong background in software engineering principles with a focus on crafting robust and maintainable solutions to challenging problems
- Background in LLVM code generation including instruction scheduling, software pipelining, register allocation, GlobalISel, TableGen, LLVM IR, and Machine IR (MIR)
- Experience in developing CUDA, DirectX, OpenGL/Vulkan applications
- 5+ years of compiler code generation experience
Responsibilities
- Understand, modify, and improve an NVIDIA proprietary GPU compiler backend 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
Other
- B.S. or degree in Computer Science/Engineering or equivalent experience
- Good communication and documentation skills and self-motivated
- M.S./PhD. with significant compiler related project or thesis work preferred
- Travel requirements not specified
- US work visa sponsorship not specified
