Stories by matt_d

Fleet: Hierarchical Task-Based Abstraction for Megakernels on Multi-Die GPUs

Fuzzing with Agents? Generators Are All You Need

Kuna: Decompiler Development in the Age of Coding Agents

PyCuTe: Reference implementation and examples of the CuTe Layout

PyTorch: A Reference Language

Tracked Capabilities for Safer Agents

Attention Decode on AMD MI450 GPUs: A Gluon Kernel Optimization Guide

Don't Use Ordinary Software to Contain Software-Hacking Agents

A practical partitioner for distributed simulations on sparse dynamic domains

NVLink, NVSwitch, and All That

Continuation-Centric Computing with Arca

AutoCO: An Online Continuous Optimization System for Phase-Changing DB Workloads

Reusing Buffers in Multicore OCaml

Mohabi: Disaggregating and Sandboxing the Firefox JavaScript Engine

Automated Discovery Has No Universally Superior Harness

Drawn Apart: Remote GPU Fingerprinting

ICFP Programming Contest 2026

Optimization of SASS Stall Counts

Characterizing Metastable Faults and Failures

Every Microsecond Matters:Achieving Near SpeedOfLight Latency in GPU Collectives

1 2 3 4 5 6 7 8 9 10