GPU Compiler Engineer (m/f/d) (fixed-term Contract)

Intel Intel · Semiconductors · Munich, Germany

The role involves designing, developing, and optimizing compiler functionality for GPU architectures, with a focus on improving performance and efficiency for AI and data center applications. This includes investigating new techniques, analyzing workloads, and collaborating with cross-functional teams. The position requires strong C/C++ skills and knowledge of LLVM internals or GPU/CPU compiler technologies.

What you'd actually do

  1. Design and implement state-of-the-art compiler translations and optimizations targeting GPU architectures
  2. Investigate, prototype, and deliver new techniques to improve performance and efficiency
  3. Analyze workloads, perform benchmarking, and evaluate compiler behavior on real hardware
  4. Collaborate with cross-functional teams to enhance compiler support for AI and data center applications
  5. Contribute to improving compilation technologies aligned with modern hardware features

Skills

Required

  • C
  • C++
  • LLVM internals or GPU/CPU compiler technologies
  • modern software development tools and practices

Nice to have

  • compiler optimizations (middle-end or backend)
  • performance analysis and tuning
  • GPU architecture and programming models
  • OpenCL or OneAPI
  • heterogeneous systems and AI workloads

What the JD emphasized

  • state-of-the-art compiler translations and optimizations
  • performance and efficiency
  • AI and data center applications
  • C and C++
  • LLVM internals and/or GPU/CPU compiler technologies

Other signals

  • compiler development
  • GPU architectures
  • AI workloads