2024

• 12-31 - Triton Linear Layout: Concept
• 12-25 - Triton Compiler Development Tips

2023

• 09-26 - Leaving Google
• 04-01 - Single-node ML Runtime Foundation

2022

• 08-31 - MLIR Linalg Dialect and Patterns
• 07-31 - MLIR Vector Dialect and Patterns
• 02-20 - MLIR CodeGen Dialects for Machine Learning Compilers
• 01-08 - Compilers and IRs: LLVM IR, SPIR-V, and MLIR

2021

• 09-19 - CodeGen Performant Convolution Kernels for Mobile GPUs
• 08-21 - Android Native Library Benchmarking Pipeline for Open Source Projects
• 07-25 - GPGPU, ML Inference, and Vulkan Compute
• 07-17 - Edge/Mobile ML Inference Challenges
• 07-08 - Sampling Performance Counters from Mobile GPU Drivers
• 07-05 - Android/Linux GPU Drivers: Internals and Resources
• 06-25 - What is Vulkan Compute?

2018

• 05-12 - Shader Toolchain: HLSL in Vulkan
• 05-11 - HLSL for Vulkan: Semantic Strings and Location Numbers
• 04-24 - HLSL for Vulkan: Resources
• 04-18 - HLSL for Vulkan: Matrices