Stories by matt_d

Precision Matters in Block Scales

Agents' Last Exam

Does the Harness Matter? Lessons from Ale-Claw on Agents' Last Exam

Demystifying NVSHMEM: System-Level: Symmetric Memory, Device-Initiated Ops

Enumerating Ill-Typed Programs for Testing Type Analyzers

Agentic Memory Management for GPU Code Generation

CommBench: Can LLMs Write Correct and Efficient GPU Communication Code?

Frontier: A Discrete-Event Simulator for Modern LLM Serving

Piper: A Programmable Distributed Training System

Radix Top-K: finding the top-k elements in an array without sorting

A Case for a Simulation-Driven Exploration of Distributed GenAI Platforms

Defeat the Heap: Zero-Copy Data Movement in AXI4MLIR

Breaking the Ice: Analyzing Cold Start Latency in vLLM

An Empirical Comparison of General Context-Free Parsers

RFC: Programming Languages Course Reboot, 2026 – Shriram Krishnamurthi

CodegenBench: Can LLMs Write Efficient Code Across Architectures?

ACM SIGPLAN Programming Language Design and Implementation (PLDI) 2026

Human Judgment as a Specification

OOBdump: Relocation Oriented Programming: Arbitrary code execution in objdump -g

Inference: Turning Electricity into Intelligence – Stanford CS336 – Dan Fu [video]

1 2 3 4 5 6 7 8 9 10