Each series takes one setup from first principles down to the kernels, one episode at a time.
Series 1 · 1 episode
One model (Qwen3.8-27B, BF16), one engine (vLLM 0.31), 1x NVIDIA H100 GPU. Each episode explains one idea with a number I measured, the source file it came from, and a Jupyter notebook that reproduces it.
