TileLang Developer

Write high-performance AI compute kernels using TileLang - a tile-based programming model that bridges the gap between CUDA's low-level control and high-level abstractions.

When to Use This Skill

Use this skill when the user needs to:

• Implement custom GPU kernels for AI operations (matrix multiplication, attention mechanisms, etc.)
• Optimize performance-critical operators for modern GPUs (NVIDIA Ampere/Hopper, AMD MI300X, Ascend NPU)
• Debug TileLang code or resolve performance issues
• Port kernels across different hardware platforms
• Understand or modify existing TileLang implementations

Kernel Development Workflow

Follow these steps when writing a TileLang kernel:

Step 1: Analyze Requirements