Posts tagged with "primer"

Green Test, Red Model: Why GPU Kernels Pass Unit Tests and Fail End to End

Why a GPU kernel can pass every unit test and still break an end-to-end model run: seven gaps between a test and a model, traced through public bugs from the gfx1250 bring-up and from FlyDSL v0.3 to v0.4, and a validation ladder that closes them.

Autotune, End to End — From Triton to FlyDSL

How kernel autotuning is designed in Triton, used in aiter / quack / CuteDSL, and consumed by inference engines like SGLang — and, from all of that, how to design a real autotune path for FlyDSL. With a full HTML deep dive carrying six hand-drawn SVG plates.

From Python to Silicon — A Compiler & Arch Primer for the Working ML Engineer

You can write production ML systems for years without knowing what IR, MLIR, LLVM, ISA, or FFI actually mean. This is the patch — a bilingual primer for the undergrad-CS-but-skipped-compilers crowd, with a full HTML deep dive carrying six hand-drawn SVG plates.