Stories by matt_d

Building an Open-Source Verilog Simulator with AI: 580K Lines in 43 Days

AgentCgroup: Understanding and Controlling OS Resources of AI Agents

Equality Saturation for Circuit Synthesis and Verification

An Introduction to Folios

Perplexity Cannot Always Tell Right from Wrong

Ganak: The Making of a Versatile, High Performance Model Counter

TorchLean: Formalizing Neural Networks in Lean

Fast Autoscheduling for Sparse ML Frameworks

TENSURE: Fuzzing Sparse Tensor Compilers (Registered Report)

A Reinforcement Learning Environment for Automatic Code Optimization in MLIR

Metamorphic Testing for Infrastructure-as-Code Engines [pdf]

K-Search: LLM Kernel Generation via Co-Evolving Intrinsic World Model

Midtraining Bridges Pretraining and Posttraining Distributions

Testing "Raw" GPU Cache Latency

Hexagon-MLIR: An AI Compilation Stack for Qualcomm's NPUs

Analyzing Latency Hiding and Parallelism in an MLIR-Based AI Kernel Compiler

Argus: Automated Discovery of Test Oracles for DBMSs Using LLMs

A Decade of Docker Containers

In Pursuit of High-Fidelity GPU Kernel Benchmarking

From ASPLOS to Orbit: Unikernels Twelve Years Later

1 2 3 4 5 6 7 8 9 10