Member-only story
What a Kernel Is, and Why Everyone Is Writing New Ones
A kernel is one small program that runs on the GPU. FlashAttention, Triton and FlashInfer are all names in this layer, and all of them exist for the same reason: the arithmetic was never the expensive part.
12 min readAug 18, 2026
Part three of seven. One layer above the silicon, and the source of most of the names you have heard.
- Part 1: Start Here: The Words Everyone Uses About LLM Inference
- Part 2: Your KV Cache Is Bigger Than Your Model
- Part 4: Quantization Is Four Decisions, Not One
- Part 5: Tuning vLLM: What Every Setting Does to the Arithmetic
- Part 6: Your Second GPU Is Bought for the Cache, Not the Model
- Part 7: Your Context Length Decides What a Kernel Is Worth
Part one established that the GPU can calculate far faster than it can fetch, and part two showed how much there is to fetch. This part is about the layer where somebody actually does something about it.
It is also where nearly all the vocabulary comes from. FlashAttention, FlashInfer, Triton, CUTLASS, CUDA: every one of those is a name in this single layer, and they make…
