# mlc-ai/agentic-gpu-programming-for-mlsys architecture

> Architecture diagram of the GitHub repository mlc-ai/agentic-gpu-programming-for-mlsys, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-09-29). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/mlc-ai/agentic-gpu-programming-for-mlsys
- Repository: https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys

## Overview

This repository is a static technical book, not the GPU harness or kernel runtime it describes. Its principal reader journey is from the book’s chapter index through concepts, the concrete TIRx system, and worked optimization examples. The map reflects the README’s stated parts and capabilities; source excerpts are limited, so connections between individual chapters are shown only where the README or tree supports them.

Main entry points: [`index.md`](https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/index.md) (Chapter index), [`_static/diagrams`](https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/tree/main/_static/diagrams) (Book diagrams), [`chapters/overview.md`](https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/overview.md) (System overview), [`chapters/analysis.md`](https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/analysis.md) (Compiler analysis)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_entry["Book entry"]
  node_index["Chapter index<br/>[index.md]"]
  node_figures["Book diagrams"]
end

subgraph group_concepts["Core concepts"]
  node_overview["System overview<br/>[overview.md]"]
  node_analysis["Compiler analysis<br/>[analysis.md]"]
  node_knowledge["Kernel knowledge<br/>[knowledge.md]"]
  node_benchmarking["Benchmarking<br/>[benchmarking.md]"]
  node_agent_workflows["Agent workflows<br/>[agent-workflows.md]"]
end

subgraph group_system["TIRx system"]
  node_harness["TIRx Harness<br/>[harness.md]"]
  node_compiler_foundation["Compiler foundation"]
  node_kcoral["KCoral<br/>[kcoral.md]"]
end

subgraph group_practice["Worked practice"]
  node_running_workflows["Running workflows"]
  node_benchmark_deepdive["Benchmark server"]
  node_self_improvement["Review and improvement"]
  node_tips["Advanced tips<br/>[advanced-tips.md]"]
  node_compiler_deepdive["Compiler analysis deep dive"]
  node_trace_examples["Synchronization traces"]
end

node_reader(("Engineer reader"))

node_reader -->|"opens"| node_index
node_index -->|"orders chapters"| node_overview
node_index -->|"orders chapters"| node_analysis
node_index -->|"orders chapters"| node_knowledge
node_index -->|"orders chapters"| node_benchmarking
node_index -->|"orders chapters"| node_agent_workflows
node_index -->|"orders chapters"| node_harness
node_index -->|"orders chapters"| node_compiler_foundation
node_index -->|"orders chapters"| node_kcoral
node_reader -->|"reads"| node_running_workflows
node_reader -->|"reads"| node_benchmark_deepdive
node_reader -->|"reads"| node_self_improvement
node_reader -->|"reads"| node_tips
node_reader -->|"reads"| node_compiler_deepdive
node_reader -->|"inspects examples"| node_trace_examples
node_reader -->|"views"| node_figures

click node_index "https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/index.md"
click node_overview "https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/overview.md"
click node_analysis "https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/analysis.md"
click node_knowledge "https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/knowledge.md"
click node_benchmarking "https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/benchmarking.md"
click node_agent_workflows "https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/agent-workflows.md"
click node_harness "https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/harness.md"
click node_compiler_foundation "https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/tirx-compiler-foundation.md"
click node_kcoral "https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/kcoral.md"
click node_running_workflows "https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/running-agent-workflows.md"
click node_benchmark_deepdive "https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/benchmark-server-deepdive.md"
click node_self_improvement "https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/self-improvement.md"
click node_tips "https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/advanced-tips.md"
click node_compiler_deepdive "https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/compiler-analysis-deepdive.md"
click node_trace_examples "https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/tree/main/traces/chapter-3"
click node_figures "https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/tree/main/_static/diagrams"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_index,node_figures toneBlue
class node_overview,node_analysis,node_knowledge,node_benchmarking,node_agent_workflows toneAmber
class node_harness,node_compiler_foundation,node_kcoral toneMint
class node_running_workflows,node_benchmark_deepdive,node_self_improvement,node_tips,node_compiler_deepdive,node_trace_examples toneRose
class node_reader toneIndigo
```

## Components

### Book entry

- [Chapter index](https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/index.md): `index.md`
- [Book diagrams](https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/tree/main/_static/diagrams): `_static/diagrams`

### Core concepts

- [System overview](https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/overview.md): `chapters/overview.md`
- [Compiler analysis](https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/analysis.md): `chapters/analysis.md`
- [Kernel knowledge](https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/knowledge.md): `chapters/knowledge.md`
- [Benchmarking](https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/benchmarking.md): `chapters/benchmarking.md`
- [Agent workflows](https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/agent-workflows.md): `chapters/agent-workflows.md`

### TIRx system

- [TIRx Harness](https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/harness.md): `chapters/harness.md`
- [Compiler foundation](https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/tirx-compiler-foundation.md): `chapters/tirx-compiler-foundation.md`
- [KCoral](https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/kcoral.md): `chapters/kcoral.md`

### Worked practice

- [Running workflows](https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/running-agent-workflows.md): `chapters/running-agent-workflows.md`
- [Benchmark server](https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/benchmark-server-deepdive.md): `chapters/benchmark-server-deepdive.md`
- [Review and improvement](https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/self-improvement.md): `chapters/self-improvement.md`
- [Advanced tips](https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/advanced-tips.md): `chapters/advanced-tips.md`
- [Compiler analysis deep dive](https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/blob/main/chapters/compiler-analysis-deepdive.md): `chapters/compiler-analysis-deepdive.md`
- [Synchronization traces](https://github.com/mlc-ai/agentic-gpu-programming-for-mlsys/tree/main/traces/chapter-3): `traces/chapter-3`

### Other components

- **Engineer reader**

## Connections

- Engineer reader → Chapter index: opens
- Chapter index → System overview: orders chapters
- Chapter index → Compiler analysis: orders chapters
- Chapter index → Kernel knowledge: orders chapters
- Chapter index → Benchmarking: orders chapters
- Chapter index → Agent workflows: orders chapters
- Chapter index → TIRx Harness: orders chapters
- Chapter index → Compiler foundation: orders chapters
- Chapter index → KCoral: orders chapters
- Engineer reader → Running workflows: reads
- Engineer reader → Benchmark server: reads
- Engineer reader → Review and improvement: reads
- Engineer reader → Advanced tips: reads
- Engineer reader → Compiler analysis deep dive: reads
- Engineer reader → Synchronization traces: inspects examples
- Engineer reader → Book diagrams: views