Stories by matt_d

C++ Lifetime-End Pointer-Zap and OOTA Progress

TIRx: An Open Compiler Stack for Evolving Frontier ML Kernels

What Is A Programming Language? – Advent of Computing Episode 184

NVFP4 Blockscaled GEMM on NVIDIA RTX Pro Blackwell GPUs (SM12x)

LLVM-Snippy: An Instruction Sequence Generator. Part 1: Overview [video]

LLMs Are Complicated Now

petite-vllm Part 2: KV Cache & Paged Attention

Analyzing Bytes: Pre-Disassembly Static Binary Analysis

Terminal-Bench Challenges: long-horizon, token-intensive, single-task benchmarks

Dana Scott: Lambda Calculus, Forcing and the Foundations of Math: #14 aboutlogic [video]

SE Radio 725: Danny Yang and Sam Goldman on the Pyrefly Type Checker

Integer Quantization: Deep Dive

M* (M-Star): A Modular, Extensible, Serving System for Multimodal Models

From Minutes to Seconds: LLM-Guided Autotuning for Helion Kernels

Zigzag Decoding with AVX-512

GenDB – LLM-Powered Generative Query Engine

AI Compute Extensions (ACE) Specification

Loop Unrolling in the ML Era

System call instrumentation on Linux/x86‑64 using memory‑indirect calls, part I

Fearless Concurrency on the GPU

1 2 3 4 5 6 7 8 9 10