Tri Dao

mentions 2 type Person feed RSS

// recent coverage 2 mentions

03:23

2026-07-31

dev.to

artificial-intelligence

How a Baseten Engineer Traced 7 Years of Attention Mechanism Evolution -- From GPT-2 to Kimi K3, in Runable PyTorch

A Baseten inference engineer known as @waterloo_intern published a technical blog post titled '22,580: From GPT-2 to Kimi K3, Explained,' which has garnered 2.4 million views. The post provides runnab…

18:15

2026-06-15

dev.to

large-language-models

Fused Kernels in LLMs: Reducing Memory Bandwidth Bottlenecks Through GPU Kernel Fusion

Shrijith Venkatramana, developer of git-lrc, explains how kernel fusion reduces memory bandwidth bottlenecks in LLM inference. By combining multiple GPU operations into a single kernel, intermediate d…

// co-occurs with top 8 entities

Shrijith Venkatramana 1 git-lrc 1 FlashAttention 1 GPU 1 Baseten 1 OpenAI 1 Moonshot AI 1 Kimi K3 1

// topics top 6 topics

large language models 2 ai research 2 ai infrastructure 1 artificial intelligence 1 machine learning 1 developer tools 1