Stories by matt_d

How to Write to SSDs [pdf]

Scalable GPU Acceleration of Scalar Functions in Analytical Databases [pdf]

The agent principal-agent problem

FrontierSmith: Synthesizing Open-Ended Coding Problems at Scale

Demystifying the Silence of Correctness Bugs in PyTorch Compiler

Fork, Explore, Commit: OS Primitives for Agentic Exploration

Systematically Auditing AI Agent Benchmarks with BenchJack

mimalloc: A new, high-performance, scalable memory allocator for the modern era

Let AI Agents Write Your Serving Stack with VibeServe

TLX: Hardware-Native, Evolvable MIMW GPU Compiler for Large-Scale Production

TorchLean: Verified Neural Networks in Lean

Deterministic Fully-Static Whole-Binary Translation Without Heuristics

Dynamic Persistent Tile Scheduling w/ Cluster Launch Control (CLC) on Blackwell

VibeServe: Can AI Agents Build Bespoke LLM Serving Systems?

CCL-Bench 1.0: A Trace-Based Benchmark for LLM Infrastructure

Microbenchmark-Driven Analytical Performance Modeling Across Modern GPUs

PyTorch DevLog

VDCores: Resource Decoupled Programming and Execution for Asynchronous GPU

Aurora: A Leverage-Aware Optimizer for Rectangular Matrices

The Two Abstractions of System Design: Hide or Reduce

4 5 6 7 8 9 10 11 12 13