Compilers and MLIR: From LLVM to the Triton and XLA Kernels That Make GPUs Fast

Deep End Era. Engineering with Sankalp · Oct 10, 2026 · 56m

A slow neural-network layer becomes a question about what a compiler can see. Begin with the compiler pipeline and LLVM, then explore why machine-learning workloads need additional representations and transformations. XLA, MLIR and kernel-generation ideas connect graph-level structure to memory movement and hardware execution. Portability and verification raise the next questions: how much optimization can be shared across accelerators, and how do we know a transformed program remains correct?

Chapters

00:00:00 Cold open: the slow layer

00:01:52 What a compiler actually does

00:04:32 LLVM…