Articles by matt_d
1

The Thing We All Obviously Want (kmicinski.com)

2

ActPlane: Programmable OS-Level Policy Enforcement for Agent Harnesses (arxiv.org)

1

Fenwick trees for products mod 2ⁿ (bitmath.blogspot.com)

1

A Fake Shell for Pangenomics (cornell.edu)

1

Making Equality Saturation Usable for Developing Vectorized Compilers (acm.org)

2

Reading AI Model Compilation in MLIR Through the Lens of Formal Theories (arxiv.org)

2

Liveness Proofs in Veil, Part I: The First Step (proofsandintuitions.net)

1

ParallelKernelBench: Can LLMs write fast multi-GPU kernels? (github.com/togethercomputer)

1

LXM: Better Splittable Pseudorandom Number Generators (and Almost as Fast) [video] (youtube.com)

2

Code as Agent Harness (arxiv.org)

2

PICO: Performance Insights for Collective Operations (ieee.org)

1

An Overview of Petri Net Theory [video] (youtube.com)

1

2026 EuroLLVM Developers' Meeting Talks (youtube.com)

2

VoltanaLLM: Energy-Efficient LLM Serving (supercomputing-system-ai-lab.github.io)

2

Why Software Requirements Get Easier in an AI Economy (stng.substack.com)

1

Lifting E-Graphs: A Function Isn't a Constant (arxiv.org)

1

Inference Compute Shapes Frontier LLM Evaluation (arxiv.org)

1

Concordia: JIT-Compiled Persistent-Kernel Checkpt for Fault-Tolerant Inference (arxiv.org)

1

SIMT-Step Execution: A Flexible Operational Semantics for GPU Subgroup Behavior (arbersephirotheca.github.io)

1

NektarIR: A Domain-Specific Compiler for High-Order FE Ops on Heterogeneous HW (arxiv.org)

2

C++ Lifetime-End Pointer-Zap and OOTA Progress (kernel.org)

1

TIRx: An Open Compiler Stack for Evolving Frontier ML Kernels (apache.org)

1

What Is A Programming Language? – Advent of Computing Episode 184 (libsyn.com)

2

NVFP4 Blockscaled GEMM on NVIDIA RTX Pro Blackwell GPUs (SM12x) (colfax-intl.com)

3

LLVM-Snippy: An Instruction Sequence Generator. Part 1: Overview [video] (youtube.com)

47

LLMs Are Complicated Now (ianbarber.blog)

1

petite-vllm Part 2: KV Cache & Paged Attention (kristenmcintosh.dev)

1

Analyzing Bytes: Pre-Disassembly Static Binary Analysis (research.google)

1

Terminal-Bench Challenges: long-horizon, token-intensive, single-task benchmarks (tbench.ai)

1

Dana Scott: Lambda Calculus, Forcing and the Foundations of Math: #14 aboutlogic [video] (youtube.com)

2

SE Radio 725: Danny Yang and Sam Goldman on the Pyrefly Type Checker (se-radio.net)

9

Integer Quantization: Deep Dive (hello-fri-end.github.io)

3

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

3

From Minutes to Seconds: LLM-Guided Autotuning for Helion Kernels (pytorch.org)

1

Zigzag Decoding with AVX-512 (zeux.io)

1

GenDB – LLM-Powered Generative Query Engine (solidlao.github.io)

20

AI Compute Extensions (ACE) Specification (x86ecosystem.org)

1

Loop Unrolling in the ML Era (hiraditya.github.io)

6

System call instrumentation on Linux/x86‑64 using memory‑indirect calls, part I (humprog.org)

1

Fearless Concurrency on the GPU (arxiv.org)

1

Using Task Graph Caching to Accelerate TVM Code Generation (acm.org)

1

Google's Training Supercomputers from TPU v2 to Ironwood: Five Generations (arxiv.org)

6

The Return of Rigorous Full-System Timing Simulation (sigarch.org)

4

Language integrated LLMs as an OCaml function (recoil.org)

2

Using OxCaml to implement type-safe reference counting between OCaml and Python (janestreet.com)

2

Scalable GPU Acceleration of Scalar Functions in Analytical Databases (microsoft.com)

1

Compiling Strassen-Like Matrix Multiplication Algorithms to Fast CUDA Kernels (acm.org)

2

Programming Language Design and Implementation (PLDI) 2026 Live Streams (sigplan.org)

1

Puzzling Success of Overparameterization: Lottery Tickets or Escape Dimensions? (epfl.ch)

2

One More Type in the Tiny Type Theory (jcreedcmu.github.io)

3

A Galois Field Arithmetic Primer (tomverbeure.github.io)

1

An O(x)Caml book that runs (kcsrk.info)

4

Type Theory Forall #62 – Dependent Haskell – Vladislav Zavialov [video] (youtube.com)

4

Trip report: June 2026 ISO C++ standards meeting (Brno, Czechia) (herbsutter.com)

1

UnpredictaBench: A Benchmark for Evaluating Distributional Randomness in LLMs (arxiv.org)

1

Linear Algebra Kernels for the Age of Research (gpumode.com)

2

Latent learning: episodic memory complements parametric learning (openreview.net)

1

NEURA: A Unified and Retargetable Compilation Framework for CGRAs (acm.org)

2

System Call Stack Alignment (humprog.org)

1

Making FlashAttention-4 faster for inference (modal.com)

1

Precision Matters in Block Scales (constantinides.net)

2

Agents' Last Exam (arxiv.org)

2

Does the Harness Matter? Lessons from Ale-Claw on Agents' Last Exam (agents-last-exam.org)

1

Demystifying NVSHMEM: System-Level: Symmetric Memory, Device-Initiated Ops (arxiv.org)

1

Enumerating Ill-Typed Programs for Testing Type Analyzers (acm.org)

1

Agentic Memory Management for GPU Code Generation (ucbskyadrs.github.io)

1

CommBench: Can LLMs Write Correct and Efficient GPU Communication Code? (uccl-project.github.io)

1

Frontier: A Discrete-Event Simulator for Modern LLM Serving (github.com/netx-lab)

2

Piper: A Programmable Distributed Training System (washington.edu)

1

Piper: A Programmable Distributed Training System (arxiv.org)

1

Radix Top-K: finding the top-k elements in an array without sorting (veitner.bearblog.dev)

1

A Case for a Simulation-Driven Exploration of Distributed GenAI Platforms (acm.org)

1

Defeat the Heap: Zero-Copy Data Movement in AXI4MLIR (arxiv.org)

2

Breaking the Ice: Analyzing Cold Start Latency in vLLM (arxiv.org)

2

An Empirical Comparison of General Context-Free Parsers (arxiv.org)

1

RFC: Programming Languages Course Reboot, 2026 – Shriram Krishnamurthi (docs.google.com)

1

CodegenBench: Can LLMs Write Efficient Code Across Architectures? (arxiv.org)

1

ACM SIGPLAN Programming Language Design and Implementation (PLDI) 2026 (acm.org)

2

Human Judgment as a Specification (brownplt.org)

3

OOBdump: Relocation Oriented Programming: Arbitrary code execution in objdump -g (calif.io)

3

Inference: Turning Electricity into Intelligence – Stanford CS336 – Dan Fu [video] (youtube.com)

5

FP8 Is All You Need (Part 1): Debunking Hardware FP64 as the HPC Holy Grail (arxiv.org)

4

Modular Arithmetic Challenge (terrytao.wordpress.com)

1

The Return of Rigorous Full-System Timing Simulation (sigarch.org)

2

Co-Creator of Haskell: Functional Prog., Thinking in Types, Useless Languages [video] (youtube.com)

1

Types for more than memory safety in OxCaml – Stephen Dolan – VeTSS 2026 [video] (youtube.com)

112

The 29th International Obfuscated C Code Contest (IOCCC) 2025 Winners (ioccc.org)

3

Tensor Shapes in Pyrefly – Avik Chaudhuri – PyCon US 2026 Typing Summit [video] (youtube.com)

1

BenchEvolver: Frontier Task Synthesis via Solution-Centric Evolution (benchevolver.github.io)

2

JITDomain: Instruction-level JIT code isolation (sciencedirect.com)

2

Serving Transformers: Lessons from the Trenches – Stanford CS25 Transformers [video] (youtube.com)

1

Constrained Adaptive Rejection Sampling (arxiv.org)

2

Training an Agentic Router for Optimal Cost-Performance on SWE Tasks (appliedcompute.com)

1

Diagramming Program Values by Spatial Refinement (brownplt.org)

1

Agent Arena: Causal Evaluation of Agents in the Real World (arena.ai)

1

Can LLMs Reason Structurally? Benchmarking via the Lens of Data Structures (arxiv.org)

1

Recent improvements to the type checker – Swift Compiler (swift.org)

1

Semantic Reification: A New Paradigm for Random Program Generation (sigplan.org)

1

GPU Forecasters: Language Models as Selective Surrogates for Kernel Optimization (arxiv.org)

1

Type-Error Ablation and AI Coding Agents (arxiv.org)