Stories by matt_d

Guardians: Static verification for AI agent workflows

Fast GPU Linear Algebra via Compile Time Expression Fusion

The AI Compute Extensions (ACE) for x86 [pdf]

Finding and Understanding Bugs in FPGA Place-and-Route Engines [video]

AutoSP: Long-Context LLM Training via Compiler-Based Sequence Parallelism

Partial UDF Inlining

From Convergence to Confidence: Push-Button Verification for RDTs

Low-Compilation-Cost Register Allocation in LLVM-Based Binary Translation

AdaExplore: Search for Efficient Kernel Generation

vLLM-Compile: Bringing Compiler Optimizations to LLM Inference

Evaluating CUDA Tile for AI Workloads on Hopper and Blackwell GPUs

Compiler Testing – Part 1: Coverage-Guided Fuzzing with Grammars and LLMs

Disaggregated Serving for Hybrid SSM Models in vLLM

Great Paper: The Calculated Typer – Iowa Type Theory Commute Podcast S7 E6

Barbara Liskov, Turing Award'08: Data Abstraction, Dijkstra, Distributed Systems

Guess-Verify-Refine: Data-Aware Top-K for Sparse-Attention Decoding on Blackwell

Reimagining Kernel Generation at the PTX Layer

A Deductive System for (Hardware-Software) Contract Satisfaction Proofs

Tile Kernels: An optimized GPU kernels library written in TileLang

AMD's Zen: Coming Back from the Dead

6 7 8 9 10 11 12 13 14 15