# kevin-sodhi/amd-adv-ai architecture

> Architecture diagram of the GitHub repository kevin-sodhi/amd-adv-ai, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-11-22). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/kevin-sodhi/amd-adv-ai
- Repository: https://github.com/kevin-sodhi/amd-adv-ai

## Overview

This repository is a collection of hands-on learning materials about AI agents, LLM inference, and AMD ROCm—not a single executable application. Learners use the notebooks and PDFs as separate workshops or labs; the available evidence does not document runtime calls or data flow between them. The map preserves the distinct learning tracks and materials visible in the file tree. README contents and source excerpts were not provided, so relationships between materials are omitted rather than inferred.

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_agents["AI Agents"]
  node_agent_guide["Multi-Agent Guide<br/>[KevinSo.pdf]"]
  node_google_adk["Google ADK Tutorial"]
end

subgraph group_serving["LLM Serving"]
  node_unified_caching["Unified Caching"]
end

subgraph group_inference["Inference Optimization"]
  node_hyperloom["Hyperloom Inference Tuning"]
  node_optimization_readme["Optimization Overview<br/>[README.md]"]
  node_optimization_demo["LLM Optimization Demo"]
  node_optimization_slides["Optimization Workshop PDF"]
end

subgraph group_openclaw["OpenClaw Workshop"]
  node_openclaw_readme["OpenClaw Overview<br/>[README.md]"]
  node_openclaw_review["Code Review Workshop"]
end

subgraph group_rocm["ROCm Certification"]
  node_lab_architecture["Architecture Microbenchmarks"]
  node_lab_kernels["Kernel Tuning"]
  node_lab_libraries["PyTorch Libraries"]
  node_lab_porting["HIP Porting"]
  node_lab_performance["Performance Optimization"]
end

click node_agent_guide "https://github.com/kevin-sodhi/amd-adv-ai/blob/main/AI_Agents_201A%20Hands-On%20Guide%20to%20Multi-Agent%20Systems/KevinSo.pdf"
click node_google_adk "https://github.com/kevin-sodhi/amd-adv-ai/blob/main/AI_Agents_201A%20Hands-On%20Guide%20to%20Multi-Agent%20Systems/Tutorial_Powering_Google_ADK_on_AMD_Platform_and_Local_LLMs.ipynb"
click node_unified_caching "https://github.com/kevin-sodhi/amd-adv-ai/blob/main/Efficient%20LLM%20Serving%20at%20Scale%20with%20Unified%20Caching/Efficient%20LLM%20Serving%20at%20Scale%20with%20Unified%20Caching.ipynb"
click node_hyperloom "https://github.com/kevin-sodhi/amd-adv-ai/blob/main/Inference%20perfomance%20tuning%20with%20Ai%20agents/workshop_hyperloom_qwen3_vllm_forge_woB.ipynb"
click node_optimization_readme "https://github.com/kevin-sodhi/amd-adv-ai/blob/main/LLM-Inference-Optimization-on-AMD-GPUs/README.md"
click node_optimization_demo "https://github.com/kevin-sodhi/amd-adv-ai/blob/main/LLM-Inference-Optimization-on-AMD-GPUs/atom_llm_demo.ipynb"
click node_optimization_slides "https://github.com/kevin-sodhi/amd-adv-ai/blob/main/LLM-Inference-Optimization-on-AMD-GPUs/ATOM-AAI-WS-2026_0723.pdf"
click node_openclaw_readme "https://github.com/kevin-sodhi/amd-adv-ai/blob/main/OpenClaw-Multimodal-Agent-on-AMD/README.md"
click node_openclaw_review "https://github.com/kevin-sodhi/amd-adv-ai/blob/main/OpenClaw-Multimodal-Agent-on-AMD/Multi-Agent%20Code%20Review%20%26%20Auto-Fix%20System%20%E2%80%94%20OpenClaw%20on%20AMD%20ROCm.ipynb"
click node_lab_architecture "https://github.com/kevin-sodhi/amd-adv-ai/blob/main/Rocm-Certification/Lab_1_Architecture_Microbenchmarks.ipynb"
click node_lab_kernels "https://github.com/kevin-sodhi/amd-adv-ai/blob/main/Rocm-Certification/Lab_2_Kernel_Tuning.ipynb"
click node_lab_libraries "https://github.com/kevin-sodhi/amd-adv-ai/blob/main/Rocm-Certification/Lab_3_Libraries_PyTorch.ipynb"
click node_lab_porting "https://github.com/kevin-sodhi/amd-adv-ai/blob/main/Rocm-Certification/Lab_4_HIP_Programming_Porting.ipynb"
click node_lab_performance "https://github.com/kevin-sodhi/amd-adv-ai/blob/main/Rocm-Certification/Lab_5_Performance_Optimization.ipynb"

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_agent_guide,node_google_adk toneBlue
class node_unified_caching toneAmber
class node_hyperloom,node_optimization_readme,node_optimization_demo,node_optimization_slides toneMint
class node_openclaw_readme,node_openclaw_review toneRose
class node_lab_architecture,node_lab_kernels,node_lab_libraries,node_lab_porting,node_lab_performance toneIndigo
```

## Components

### AI Agents

- [Multi-Agent Guide](https://github.com/kevin-sodhi/amd-adv-ai/blob/main/AI_Agents_201A%20Hands-On%20Guide%20to%20Multi-Agent%20Systems/KevinSo.pdf): `AI_Agents_201A Hands-On Guide to Multi-Agent Systems/KevinSo.pdf`
- [Google ADK Tutorial](https://github.com/kevin-sodhi/amd-adv-ai/blob/main/AI_Agents_201A%20Hands-On%20Guide%20to%20Multi-Agent%20Systems/Tutorial_Powering_Google_ADK_on_AMD_Platform_and_Local_LLMs.ipynb): `AI_Agents_201A Hands-On Guide to Multi-Agent Systems/Tutorial_Powering_Google_ADK_on_AMD_Platform_and_Local_LLMs.ipynb`

### LLM Serving

- [Unified Caching](https://github.com/kevin-sodhi/amd-adv-ai/blob/main/Efficient%20LLM%20Serving%20at%20Scale%20with%20Unified%20Caching/Efficient%20LLM%20Serving%20at%20Scale%20with%20Unified%20Caching.ipynb): `Efficient LLM Serving at Scale with Unified Caching/Efficient LLM Serving at Scale with Unified Caching.ipynb`

### Inference Optimization

- [Hyperloom Inference Tuning](https://github.com/kevin-sodhi/amd-adv-ai/blob/main/Inference%20perfomance%20tuning%20with%20Ai%20agents/workshop_hyperloom_qwen3_vllm_forge_woB.ipynb): `Inference perfomance tuning with Ai agents/workshop_hyperloom_qwen3_vllm_forge_woB.ipynb`
- [Optimization Overview](https://github.com/kevin-sodhi/amd-adv-ai/blob/main/LLM-Inference-Optimization-on-AMD-GPUs/README.md): `LLM-Inference-Optimization-on-AMD-GPUs/README.md`
- [LLM Optimization Demo](https://github.com/kevin-sodhi/amd-adv-ai/blob/main/LLM-Inference-Optimization-on-AMD-GPUs/atom_llm_demo.ipynb): `LLM-Inference-Optimization-on-AMD-GPUs/atom_llm_demo.ipynb`
- [Optimization Workshop PDF](https://github.com/kevin-sodhi/amd-adv-ai/blob/main/LLM-Inference-Optimization-on-AMD-GPUs/ATOM-AAI-WS-2026_0723.pdf): `LLM-Inference-Optimization-on-AMD-GPUs/ATOM-AAI-WS-2026_0723.pdf`

### OpenClaw Workshop

- [OpenClaw Overview](https://github.com/kevin-sodhi/amd-adv-ai/blob/main/OpenClaw-Multimodal-Agent-on-AMD/README.md): `OpenClaw-Multimodal-Agent-on-AMD/README.md`
- [Code Review Workshop](https://github.com/kevin-sodhi/amd-adv-ai/blob/main/OpenClaw-Multimodal-Agent-on-AMD/Multi-Agent%20Code%20Review%20%26%20Auto-Fix%20System%20%E2%80%94%20OpenClaw%20on%20AMD%20ROCm.ipynb): `OpenClaw-Multimodal-Agent-on-AMD/Multi-Agent Code Review & Auto-Fix System — OpenClaw on AMD ROCm.ipynb`

### ROCm Certification

- [Architecture Microbenchmarks](https://github.com/kevin-sodhi/amd-adv-ai/blob/main/Rocm-Certification/Lab_1_Architecture_Microbenchmarks.ipynb): `Rocm-Certification/Lab_1_Architecture_Microbenchmarks.ipynb`
- [Kernel Tuning](https://github.com/kevin-sodhi/amd-adv-ai/blob/main/Rocm-Certification/Lab_2_Kernel_Tuning.ipynb): `Rocm-Certification/Lab_2_Kernel_Tuning.ipynb`
- [PyTorch Libraries](https://github.com/kevin-sodhi/amd-adv-ai/blob/main/Rocm-Certification/Lab_3_Libraries_PyTorch.ipynb): `Rocm-Certification/Lab_3_Libraries_PyTorch.ipynb`
- [HIP Porting](https://github.com/kevin-sodhi/amd-adv-ai/blob/main/Rocm-Certification/Lab_4_HIP_Programming_Porting.ipynb): `Rocm-Certification/Lab_4_HIP_Programming_Porting.ipynb`
- [Performance Optimization](https://github.com/kevin-sodhi/amd-adv-ai/blob/main/Rocm-Certification/Lab_5_Performance_Optimization.ipynb): `Rocm-Certification/Lab_5_Performance_Optimization.ipynb`
