writing
Tiled Thoughts — a verbose debug build
Compiler passes, ML systems, and the occasional detour. Long-form notes from the path between code and silicon.
2026
2025
- Nov 22
Can I Beat Clang’s Auto-Vectorizer on Apple Silicon? A SAXPY Case Study ↗
Hand-written NEON intrinsics, loop unrolling, and a spoiler you can probably guess.
- Nov 18
Quantization in MLIR: Types, Scales, and Where to Put the q ↗
Everyone agrees it matters; the details stay fuzzy. Here are the details.
- Nov 15
How GPUs Talk: A Practical Guide to Multi-GPU Training and Communication ↗
Collectives, interconnects, and the communication patterns behind efficient distributed training.
- Nov 9
Data Structure and Iterator Kung Fu in LLVM ↗
Zero-copy views and safe mutation loops for faster LLVM passes.
- Nov 9
Demystifying GPU Terminology: A Compiler Engineer’s Field Guide ↗
SIMD, warps, occupancy, coalescing, shared memory, spills and matrix units — mapped to real compiler decisions.
- Nov 8
How Rewriting works in MLIR ↗
How MLIR's pattern rewriting works, and why it makes optimizations composable.
- Nov 4
Booting Up: A Verbose Debug Build of Life and Compilers ↗
Compiler passes, ML systems, and life — debug logs from the path between code and silicon.