Articles by matt_d
2

Classifying Capabilities (Extended Version) (arxiv.org)

1

TensorLift: Auto Extraction of ISA Semantics from Accelerator RTL via MLIR (arxiv.org)

1

A Revised Haskell 2010 Language Report (haskell.org)

1

User awareness in frontier models: Who's asking shifts what models say (transluce.org)

1

OBASE: Object-Based Address-Space Engineering to Improve Memory Tiering (usenix.org)

1

Domain-specific hyperspecialization: Winning the SC26 SAT track with LymphoSAT (c.mov)

1

Benchmarking the Residual: What Long-Horizon Evals Add Beyond Short-Task Perf (arxiv.org)

1

Automatically Retrofitting JIT Compilers (infoq.com)

1

tla-rs: IronFleet and AutoMan in Verus (github.com/stonysystems)

1

Introducing Dogwood: runtime verification for AI agents (amazon.com)

1

Beyond the Library: An Agentic Framework for Autoformalizing Research Math (arxiv.org)

1

Modeling LLM Performance from First Principles (sidbabu.com)

1

LettuceDetect v2 in Semantic Router: Gen. Hallucination Detection vLLM Endpoint (vllm-sr.ai)

1

Continual Learning via Real-Time RL for Agents (rllm-project.com)

3

Smalltalk Report from 1991 to 1996 (github.com/ducasse)

1

The State of MPI: Current Standard and Future Plans (fau.de)

3

A Primer to ML Compilers (aramesh10.github.io)

2

Inside Kimi K3's AgentENV: Can It Really Fork in 100 ms? (gensee.ai)

2

Enabling the next iteration of the borrow checker on nightly (rust-lang.org)

1

Agentic Coding in the Wild: Characterizing GitHub Copilot Traces at Production (arxiv.org)

1

Certified Program Synthesis with a Multi-Modal Verifier [pdf] (verse-lab.org)

1

Compiler-Grounded Hierarchical Diagnosis for LLM Triton Kernel Optimization (arxiv.org)

1

NIIN: NVSHMEM Implemented In NCCL (github.com/nvidia)

1

ACM SIGOPS Symposium on Operating Systems Principles SOSP'26 Papers & Preprints (pchaigno.github.io)

2

Reliability Lessons from SQLite – Richard Hipp – SSW 2026 [video] (youtube.com)

8

Stanford CS329A: Self-Improving AI Agents (youtube.com)

1

Demystifying Deep Learning Compiler Front End Bugs: An LLM-Aided Empirical Study (arxiv.org)

1

Towards Designing an Execution Control System with Metastability Resilience (muratbuffalo.blogspot.com)

1

AdaMAST: Adaptive Failure Taxonomies for Improving LLM Agents (multi-agent-systems-failure-taxonomy.gi...

2

k-Coloring is Faster than Computing the Chromatic Number (arxiv.org)

2

x86 AMX/ACE with >8 tiles (kernel.org)

1

RSI: Recursive Self-Improvement (github.com/zartbot)

1

Specula: Scaling formal specs for autonomous model checking of system code (arxiv.org)

2

NVIDIA-labs OO Agents: Native Python Object-Oriented Agents (github.com/nvidia-nemo)

1

At-the-Roofline Sparse Tensor Contractions on Vector Processors for Inference (arxiv.org)

2

Characterizing Warp Divergence from Pascal to Blackwell (arxiv.org)

2

Why Higher-Order Logic Is a Good Foundation for Deep Verification (sequent.inc)

1

Fleet: Hierarchical Task-Based Abstraction for Megakernels on Multi-Die GPUs (arxiv.org)

1

Fuzzing with Agents? Generators Are All You Need (arxiv.org)

22

Kuna: Decompiler Development in the Age of Coding Agents (noelo.org)

1

PyCuTe: Reference implementation and examples of the CuTe Layout (github.com/nvlabs)

24

PyTorch: A Reference Language (pytorch.org)

2

Tracked Capabilities for Safer Agents (martinodersky.substack.com)

7

Attention Decode on AMD MI450 GPUs: A Gluon Kernel Optimization Guide (amd.com)

1

Don't Use Ordinary Software to Contain Software-Hacking Agents (verse.systems)

1

A practical partitioner for distributed simulations on sparse dynamic domains (acm.org)

1

NVLink, NVSwitch, and All That (doubleword.ai)

1

Continuation-Centric Computing with Arca (usenix.org)

2

AutoCO: An Online Continuous Optimization System for Phase-Changing DB Workloads (acm.org)

1

Reusing Buffers in Multicore OCaml (gazagnaire.org)

3

Mohabi: Disaggregating and Sandboxing the Firefox JavaScript Engine (usenix.org)

1

Automated Discovery Has No Universally Superior Harness (arxiv.org)

2

Drawn Apart: Remote GPU Fingerprinting (bgu.ac.il)

2

ICFP Programming Contest 2026 (icfpcontest2026.com)

1

Optimization of SASS Stall Counts (redplait.blogspot.com)

1

Characterizing Metastable Faults and Failures (muratbuffalo.blogspot.com)

1

Every Microsecond Matters:Achieving Near SpeedOfLight Latency in GPU Collectives (arxiv.org)

12

An Empirical Study: AI Agent Rules Need Context and Layered Enforcement (eunomia.dev)

4

SPIR-V on ROCm: A Portable IR for AMD GPUs (amd.com)

1

GEMM Performance Measurement Methodology Guidelines (nvidia.com)

1

Faster Algorithms for Structured Matrix Multiplication via Flip Graph Search (acm.org)

2

AI for Systems is "AGI-Complete" (acm.org)

1

Controlling Reasoning Effort in LLMs (sebastianraschka.com)

3

Making TLA+ and x86 Kiss via Z3Py (philipzucker.com)

2

Irreducible Loops (maskray.me)

7

Fleet: Hierarchical Task-Based Abstraction for Megakernels on Multi-Die GPUs (arxiv.org)

4

Accelerating Block Low-Rank Foundation Model Inference on MemoryConstrained GPUs (acm.org)

2

Locality-Aware Automatic Differentiation on the GPU for Mesh-Based Computations (acm.org)

1

Triton Plugin Extensions: Enabling TLX and Custom Compiler Passes Out of the Box (pytorch.org)

1

What If the Harness Comes Before Pretraining? A Data Flywheel Perspective (hanchenli.github.io)

2

Mimesys: Turn Resource Usage Traces into Executable Workloads (usenix.org)

2

System call instrumentation on Linux/x86-64 using memory-indirect calls (humprog.org)

2

Bidirectional Elaborators à la Carte (arxiv.org)

2

CTA-Pipelining: A Latency-Oriented Spatial Scaling Method for Multi-GPU Systems (arxiv.org)

1

AI Model Co-Design: Hardware-Friendly LLM Design (nvidia.com)

1

Towards Free Normalization: Fusing Normalization into GEMM and Attention Kernels (pytorch.org)

1

Negotiating AI in Open Source Software Communities: A Case Study of LLVM Project (gu.se)

2

Reducing HBM Bottlenecks in JAX-Based LLM Training with Host Offloading (nvidia.com)

2

Compiler Testing – Part 2: Metamorphic Testing with Verified Identities (nowarp.io)

6

What Every Python Developer Should Know About the CPython ABI (quansight.org)

1

A CIRCT (Circuit IR Compilers and Tools) Project Tutorial (samuelcoward.co.uk)

1

Lifting Terms: Making Well Scoped Syntax Dumber (philipzucker.com)

1

Sheaves in Haskell (tweag.io)

1

Writing static checks to an unsuspecting library with Liquid Haskell (tweag.io)

1

Harvesting Sub-Microsecond CXL Memory Stalls with LiteSwitch (usenix.org)

3

Formally Verifying AI-Generated GPU Kernels (gimletlabs.ai)

2

Occupancy Math on the AMD MI355X GPU (CDNA4): A From-First-Principles Guide (amd.com)

3

Intelligence Is Free, Now What? Data Systems For, Of, and by Agents (bair.berkeley.edu)

2

Binvariants: Register-Level Invaraint-Guided Fuzzing for Binaries (github.com/futureslab)

2

PuzzleMoE: Efficient Compression of Large Mixture-of-Experts Models (supercomputing-system-ai-lab.github.io)

2

Ekka: Automated Diagnosis of Silent Errors in LLM Inference (washington.edu)

2

Architecture 2.0: Designing AI-Assisted Loops for Computing Systems (mlsysbook.ai)

2

Beyond Prediction: Tail-Aware Scheduling for LLM Inference (yl3469.github.io)

2

SOLAR: AI-Powered Speed-of-Light Performance Analysis (arxiv.org)

2

The Verification Horizon: No Silver Bullet for Coding Agent Rewards (arxiv.org)

4

Binary Coverage the Wrong Way (redvice.org)

2

Stealing 50 Years of Database Ideas for AI Agents (onewill.ai)

1

Empirical Computation: Prompting versus Programming [pdf] (mboehme.github.io)

1

Programming Language Design and Implementation in the Era of Machine Learning [video] (youtube.com)

63

Senior SWE-Bench: open-source benchmark that assesses agents as senior engineers (snorkel.ai)