Use software simulators and architectural modeling tools to analyze key graphics and ML workloads and identify performance and power bottlenecks.
Collaborate with hardware and software architects to define and guide HW/SW features that improve performance per watt, performance per area, and total system efficiency.
Rapidly prototype architecture concepts through shader development, algorithm implementation, and performance modeling.
Develop optimized kernels and benchmark them against competitive solutions to surface improvement opportunities.
Build and enhance modeling and simulation tools to enable fast and accurate architecture exploration.
Create algorithms and analysis workflows to uncover and eliminate systemic performance limiters across the pipeline.
Model and evaluate proposed hardware and software features to influence product capabilities, cost, and overall performance.
requirements-expected :
Expertise in C++ and Python
Experience with GFX architecture, drivers, or compilers
Hands-on work with modeling and simulation tools
Familiarity with C++ build systems such as CMake, Meson, or Bazel
Experience with graphics APIs such as Direct3D, Vulkan, or OpenGL
Experience with CUDA or ROCm development
Understanding of the graphics rendering pipeline
Knowledge of modern ML architectures, including Transformers and FlashAttention
Strong problem-solving skills and the ability to collaborate in a large, multi‑geo engineering organization
Bachelor’s or Master’s degree in Computer Science, Computer Engineering, Electrical Engineering, or a related field.
