AI Systems Architect – GPU Software/hardware Co-design

AMD AMD · Semiconductors · San Jose, CA · Engineering

AMD is seeking an AI Systems Architect to work on software and hardware co-design for GPU AI compute solutions. The role involves co-optimizing AI workloads, contributing to open-source AI frameworks, and designing new technologies for future compute platforms. Experience with AI infrastructure, kernel optimization, and GPU architecture is preferred.

What you'd actually do

  1. Collaborate with AMD architecture engineers on GPU software and hardware design decisions.
  2. Co-optimize AI workloads on current and next generation GPUs.
  3. Contribute to client projects and open-source AI frameworks, driving features and adoption.
  4. Design and develop new technologies for future compute platforms.
  5. Participate in new hardware bring-up and validation.

Skills

Required

  • GPU software and hardware design
  • AI workload co-optimization
  • open-source AI frameworks
  • new technology design
  • hardware bring-up and validation

Nice to have

  • Deep AI infrastructure experience with open-source frameworks (e.g., SGLang, vLLM)
  • Strong kernel optimization skills using DSLs and HIP (or CUDA), plus PTX/SASS equivalents
  • Hands-on knowledge of modern GPU architecture
  • Demonstrated open-source contributions on GitHub

What the JD emphasized

  • end-to-end development of startup/GPU AI hardware
  • programming models
  • GPU microarchitectures
  • tapeout
  • bring-up

Other signals

  • AI compute solutions
  • AI workloads
  • future compute platforms