HNHacker News
TopNewBestAskShowJobs

matt_d

22,535 karma · joined April 21, 2014

submissionscomments

VibeSys on Trainium: Agents Build/Tune Qwen3.8 Serving Engines on AWS Trainium2

syfi.cs.washington.edu·2 pts·matt_d·
0

Rethinking Abstractions Towards a DSL and Compiler Foundation in the Agentic Era

tvm.apache.org·1 pts·matt_d·
0

A Heart Means Recursion: Exploring Visual Meaning and Communication in VLMs

arxiv.org·3 pts·matt_d·
0

Adversarial Speculative Reasoning Improves Long Horizon Agents

alphaxiv.org·2 pts·matt_d·
0

Larger Tokenizers Suffer More Multi-Epoch Regret

wenhaochai.com·3 pts·matt_d·
0

Reasoning Models Can Control Their CoT When They Tell Themselves To

arxiv.org·3 pts·matt_d·
1

The lost malloc()

napcakes.nekoweb.org·161 pts·matt_d·
78

Multi-Token Prediction Done Right

arxiv.org·13 pts·matt_d·
4

SpecPilot: Autonomous Algorithm Discovery and Systems Optimization for SpecDec

arxiv.org·1 pts·matt_d·
0

Hinted Scaling Laws

arxiv.org·2 pts·matt_d·
0

TLAPS-Bench: Formally Proving Correctness with TLA+ Proof System

arxiv.org·1 pts·matt_d·
0

Superscalar Loop Processors

arxiv.org·1 pts·matt_d·
0

XLA from the Trenches: Correctness on TPU Workloads at Scale

arxiv.org·1 pts·matt_d·
0

h1: A Taste of the Bitter Lesson in Automated Theorem Proving

arxiv.org·2 pts·matt_d·
0

Internals of FP8 MMA on Hopper

simonveitner.substack.com·1 pts·matt_d·
0

Making CUDA Checkpointing 15x Faster on a B300: A Worklog

maknee.github.io·2 pts·matt_d·
0

From GPU to TPU with Coding Agents: What Works, What Fails, and What It Costs?

arxiv.org·1 pts·matt_d·
0

TEMPO: A Tag-Based Framework for Efficient Memory Ordering

arxiv.org·3 pts·matt_d·
0

How Well Do Agents Perform on DevOps Tasks?

cloudopsbench.ai·1 pts·matt_d·
0

CEPO: General Framework for Improving Prompt Optimization in Multi-Agent Systems

arxiv.org·2 pts·matt_d·
0

DMA-Latte: Expanding the Reach of DMA Offloads to Latency-Bound ML Communication

arxiv.org·1 pts·matt_d·
0

What does "verified memory safe" mean?

focs-lab.comp.nus.edu.sg·1 pts·matt_d·
0

A Self-Defining Operator for Distributed Applications

arxiv.org·1 pts·matt_d·
0

Highly Optimizing Ahead-of-Time Compilers for Dynamic Languages [pdf]

olimel.net·1 pts·matt_d·
0

When AI Agrees Too Much: Manipulating Automated Code Review via Bias Injection [pdf]

dimitro.gr·2 pts·matt_d·
0

FrontierCS 2: A Benchmark for Computer Science R&D

frontier-cs.org·1 pts·matt_d·
0

TorchTPU: Hybrid Eager/Compiled Architecture for Native PyTorch Execution on TPU

arxiv.org·1 pts·matt_d·
1

Language-Based Agent Control

arxiv.org·2 pts·matt_d·
0

Reactive Optimizers: rewrite the optimizer's source code during training

xingjianbai.com·32 pts·matt_d·
3

Agency: Open-source platform for debugging, profiling, and analyzing AI agents

github.com·4 pts·matt_d·
1
← PreviousPage 3 of 34Next →