GPU Programming

Knows GPU computing basics: CUDA cores, memory hierarchy, kernel execution. Understands how PyTorch uses GPU for tensor operations and runs simple LLM inference on GPU.

Understands the fundamentals of GPU Programming. Applies basic practices in daily work. Follows recommendations from the team and documentation.

Understands GPU programming in Unreal: Nanite, Lumen, Virtual Shadow Maps. Uses GPU Visualizer for rendering performance analysis. Understands draw call optimization through instancing.

Independently optimizes GPU utilization for LLM: mixed precision (FP16/BF16), memory management, CUDA stream optimization. Profiles GPU workloads using nvidia-smi and PyTorch profiler.

Develops GPU-optimized solutions in Unity: Compute Shaders for parallel computation, indirect rendering, custom SRP passes. Profiles GPU performance with RenderDoc and Unity GPU Profiler.

Develops GPU-optimized solutions in Unreal: custom compute shaders via HLSL, Niagara GPU particles, custom render passes via RDG (Render Dependency Graph).

Designs custom CUDA kernels for LLM inference optimization: fused attention, quantized matmul, custom activation functions. Uses Triton for writing high-performance GPU operations.

Designs GPU architecture for Unity projects: custom render pipelines, GPU-driven rendering, async compute. Optimizes GPU memory bandwidth and occupancy for mobile and VR platforms.

Designs GPU architecture for AAA Unreal projects: custom rendering features, GPU-driven pipeline, async compute. Optimizes for specific GPU architectures (AMD RDNA, NVIDIA Ada).

Defines GPU programming standards for the LLM team. Establishes best practices for CUDA optimization, memory management, multi-GPU coordination. Conducts GPU-intensive code reviews.

Defines GPU programming strategy for Unity studio. Establishes render pipeline standards, GPU performance budgets. Coordinates graphics programmers and tech artists.

Defines GPU programming strategy for the Unreal studio. Establishes rendering performance budgets. Coordinates graphics programmers and tech artists.

Shapes GPU optimization strategy for organizational LLM infrastructure. Defines approaches to custom kernel development, hardware-specific optimization (A100, H100), and GPU resource planning.

Defines graphics technology strategy at studio level. Evaluates new GPU capabilities (ray tracing, mesh shaders, variable rate shading). Shapes rendering roadmap.

Defines graphics technology strategy at studio level. Evaluates ray tracing, mesh shaders, neural rendering for Unreal. Shapes rendering roadmap.