Core Maintainers

• @Chivier (Yeqi Huang)
• @anyin233 (Yanwei Ye)
• @luomai (Luo Mai)
• @SecretSettler (Yinsicheng Jiang)
• @lausannel (Zhan Lu)
• @Andrewxu313 (Tairan Xu)
• @TKONIY (Yangshen Deng)
• @PeterSH6 (Guangming Sheng)
• @CongjieHe (Congjie He)
• @DD-DuDa (Dayou Du)

Update the build system

• Migrate to mdbook (@Chivier)
• Contribution Guide - How to update/add a chapter and maintain both EN/CN versions (@Chivier )
• Update website frontpage (differ from v1 and v2) (@anyin233 )
• Have new Skills for drawing a consistent style of the figures (@Chivier @anyin233 )

Update book content

• Preface — Background of the book, target readers, and evolution of ML systems (@luomai )
• 1 Introduction — Overview of machine learning system architecture and technology stack (@luomai )

Programming the System

• 2 Programming Interfaces and Computational Graphs — Tensor abstraction, automatic differentiation, graph representation and execution (@Chivier @SecretSettler )
• 3 AI Accelerators and Programming — GPU architecture and CUDA / Triton / CUTLASS programming models (@Chivier @CongjieHe @DD-DuDa)
• 4 AI Compilers and Runtime Systems — IR design, graph optimization, kernel generation, and runtime execution (@anyin233 )

Scaling the System

• 5 Data Processing Systems — Data loading, data pipelines, and distributed data processing (@Chivier )
• 6 Training Systems — Single-node and distributed training, parallelism strategies, and training optimization (@luomai )
• 7 Model Serving — Inference optimization, online serving, and model management (@lausannel @Andrewxu313 )
• 8 RL Systems — Reinforcement learning pipelines, environment interaction, and RL system design (@TKONIY @PeterSH6)
• 9 Large-scale GPU Cluster Management — GPU scheduling, resource management, and large-scale training infrastructure (@SecretSettler @anyin233)