gpu 5 When XLA Isn't Enough: Pallas, Mosaic, and Triton Jul 29, 2026 A Tour of XLA: The Compiler Beneath JAX, TensorFlow, and PyTorch Jul 23, 2026 Triton: The Compiler That Pretends to Be a Library Jul 19, 2026 Anatomy of a CUDA Binary Jul 14, 2026 The Anatomy of an Instruction Pipeline Hazard Jul 12, 2026