Tuan's Blog

A high-performance technical blog dedicated to the frontiers of AI Systems, Generative AI, Efficient AI, and Machine Learning Theory.

Built with MkDocs and Material for MkDocs.

🌟 Key Features

• Responsive Technology: Clean, modern UI with "Outfit" typography and glassmorphism headers.
• Engineering-Focused: First-class support for LaTeX math ($$ E=mc^2 $$), Python code highlighting, and request/response examples.
• Automated Workflow: Fully CI/CD integrated with GitHub Actions.
• Performance: Lightning-fast static site generation.

📂 Project Structure

mtuann.blog/
├── docs/                   # Content Source
│   ├── index.md            # Landing Page
│   ├── ai-systems/         # System Engineering & Infrastructure
│   ├── genai/              # LLMs, Diffusion, & Reasoning
│   ├── efficient-ai/       # Quantization, Pruning, & Edge AI
│   ├── trustworthy-ai/     # Safety, Alignment, & Interpretability
│   ├── paper-reviews/      # Arxiv Analysis & Summaries
│   ├── ml/                 # Foundations & Theory
│   ├── stylesheets/        # Custom CSS (Glassmorphism, etc.)
│   └── snippets/           # Reusable code blocks
├── mkdocs.yml              # Main configuration
├── requirements.txt        # Python dependencies
└── .github/workflows/      # Deployment pipelines

🚀 Getting Started

Prerequisites

• Python 3.10+
• uv (Recommended) or pip

Installation

We use uv for ultra-fast environment management.

# 1. Clone the repo
git clone https://github.com/mtuann/blog.git
cd blog

# 2. Create virtual env
uv venv

# 3. Activate
source .venv/bin/activate

# 4. Install dependencies
uv pip install -r requirements.txt

Local Development

Start the hot-reloading development server:

mkdocs serve

Visit http://127.0.0.1:8000 in your browser.

✍️ Writing Content

This blog uses extended Markdown features.

Math (LaTeX)

$$
\text{Attention}(Q, K, V) = \text{softmax}\left(\frac{QK^T}{\sqrt{d_k}}\right)V
$$

Icons & Emojis

Use Material Design icons directly in text:

• :material-brain: -> 🧠 (styled icon)
• :material-server-network: -> 🖥️ (server icon)

Admonitions

!!! tip "Optimization Tip"
    Use FlashAttention-2 to reduce memory footprint by quadratic factors.

🚢 Deployment

Automated (Recommended): This requires no manual effort.

1. Push changes to main.
2. GitHub Actions (.github/workflows/deploy.yml) builds and deploys to gh-pages automatically.

Manual:

mkdocs gh-deploy

📜 License

MIT © 2026 Tuan Nguyen