feat: add mdbook setup with GitHub Pages deploy

Mirrors the Agents-That-Think-and-Self-Improve mdbook layout: chapters
moved into src/ (with a chapters -> src/chapters symlink so existing
links keep working), SUMMARY.md, mermaid preprocessor, navy theme, and
a Pages workflow that deploys on push to main.

Author: Rushit

.github/workflows/pages.yml

@@ -0,0 +1,57 @@
+name: Deploy mdBook to GitHub Pages
+
+on:
+  push:
+    branches: [main]
+  workflow_dispatch:
+
+permissions:
+  contents: read
+  pages: write
+  id-token: write
+
+concurrency:
+  group: pages
+  cancel-in-progress: false
+
+env:
+  MDBOOK_VERSION: 0.4.40
+  MDBOOK_MERMAID_VERSION: 0.14.0
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Install mdBook + mdbook-mermaid
+        run: |
+          mkdir -p "$HOME/.local/bin"
+          curl -sSL "https://github.com/rust-lang/mdBook/releases/download/v${MDBOOK_VERSION}/mdbook-v${MDBOOK_VERSION}-x86_64-unknown-linux-gnu.tar.gz" \
+            | tar -xz -C "$HOME/.local/bin"
+          curl -sSL "https://github.com/badboy/mdbook-mermaid/releases/download/v${MDBOOK_MERMAID_VERSION}/mdbook-mermaid-v${MDBOOK_MERMAID_VERSION}-x86_64-unknown-linux-gnu.tar.gz" \
+            | tar -xz -C "$HOME/.local/bin"
+          echo "$HOME/.local/bin" >> "$GITHUB_PATH"
+
+      - name: Build book
+        run: mdbook build
+
+      - name: Setup Pages
+        uses: actions/configure-pages@v5
+
+      - name: Upload artifact
+        uses: actions/upload-pages-artifact@v3
+        with:
+          path: ./book
+
+  deploy:
+    needs: build
+    runs-on: ubuntu-latest
+    environment:
+      name: github-pages
+      url: ${{ steps.deployment.outputs.page_url }}
+    steps:
+      - name: Deploy to GitHub Pages
+        id: deployment
+        uses: actions/deploy-pages@v4

book.toml

@@ -0,0 +1,36 @@
+[book]
+title = "The Builder's Handbook: LLM Inference from First Principles"
+authors = ["TBH Books"]
+description = "Learn how LLM inference works at scale — by building one from scratch."
+language = "en"
+multilingual = false
+src = "src"
+
+[output.html]
+default-theme = "navy"
+preferred-dark-theme = "navy"
+git-repository-url = "https://github.com/tbhbooks/LLM-Inference-from-First-Principles"
+edit-url-template = "https://github.com/tbhbooks/LLM-Inference-from-First-Principles/edit/main/{path}"
+site-url = "/LLM-Inference-from-First-Principles/"
+no-section-label = false
+smart-punctuation = true
+additional-js = ["mermaid.min.js", "mermaid-init.js"]
+
+[output.html.fold]
+enable = true
+level = 1
+
+[output.html.search]
+enable = true
+limit-results = 30
+use-boolean-and = true
+boost-title = 2
+boost-hierarchy = 1
+boost-paragraph = 1
+expand = true
+heading-split-level = 3
+
+[preprocessor]
+
+[preprocessor.mermaid]
+command = "mdbook-mermaid"

chapters

@@ -0,0 +1 @@
+src/chapters

mermaid-init.js

@@ -0,0 +1,35 @@
+(() => {
+    const darkThemes = ['ayu', 'navy', 'coal'];
+    const lightThemes = ['light', 'rust'];
+
+    const classList = document.getElementsByTagName('html')[0].classList;
+
+    let lastThemeWasLight = true;
+    for (const cssClass of classList) {
+        if (darkThemes.includes(cssClass)) {
+            lastThemeWasLight = false;
+            break;
+        }
+    }
+
+    const theme = lastThemeWasLight ? 'default' : 'dark';
+    mermaid.initialize({ startOnLoad: true, theme });
+
+    // Simplest way to make mermaid re-render the diagrams in the new theme is via refreshing the page
+
+    for (const darkTheme of darkThemes) {
+        document.getElementById(darkTheme).addEventListener('click', () => {
+            if (lastThemeWasLight) {
+                window.location.reload();
+            }
+        });
+    }
+
+    for (const lightTheme of lightThemes) {
+        document.getElementById(lightTheme).addEventListener('click', () => {
+            if (!lastThemeWasLight) {
+                window.location.reload();
+            }
+        });
+    }
+})();

mermaid.min.js

@@ -0,0 +1 @@
+../README.md

src/README.md

@@ -0,0 +1,43 @@
+# Summary
+
+[Introduction](README.md)
+[Preface](chapters/preface.md)
+
+# Getting Started
+
+- [Chapter 0: Setup](chapters/ch00-setup.md)
+
+# Part I: Foundations
+
+- [Chapter 1: The LLM Inference Problem](chapters/ch01-the-llm-inference-problem.md)
+- [Chapter 2: vLLM Architecture Overview](chapters/ch02-vllm-architecture-overview.md)
+- [Chapter 3: Setting Up the Project](chapters/ch03-setting-up-the-project.md)
+
+# Part II: MVP — Single-Request Inference
+
+- [Chapter 4: GPT-2 from Scratch](chapters/ch04-gpt2-from-scratch.md)
+- [Chapter 5: The Building Blocks](chapters/ch05-the-building-blocks.md)
+- [Chapter 6: Where the Model Learns to Look Back](chapters/ch06-where-the-model-learns-to-look-back.md)
+- [Chapter 7: The Skeleton Speaks](chapters/ch07-the-skeleton-speaks.md)
+- [Chapter 8: Fit and Finish](chapters/ch08-fit-and-finish.md)
+
+# Part III: Core vLLM — Memory & Scheduling
+
+- [Chapter 9: The Memory Problem](chapters/ch09-the-memory-problem.md)
+- [Chapter 10: Paged Attention](chapters/ch10-paged-attention.md)
+- [Chapter 11: Continuous Batching](chapters/ch11-continuous-batching.md)
+- [Chapter 12: The Scheduler](chapters/ch12-the-scheduler.md)
+- [Chapter 13: The Engine Loop](chapters/ch13-the-engine-loop.md)
+
+# Part IV: Production
+
+- [Chapter 14: Sampling Strategies](chapters/ch14-sampling-strategies.md)
+- [Chapter 15: Building the API Server](chapters/ch15-building-the-api-server.md)
+- [Chapter 16: Prefix Caching](chapters/ch16-prefix-caching.md)
+- [Chapter 17: Speculative Decoding](chapters/ch17-speculative-decoding.md)
+- [Chapter 18: Structured Output](chapters/ch18-structured-output.md)
+- [Chapter 19: Parallelism](chapters/ch19-parallelism.md)
+
+# Part V: Further
+
+- [Chapter 20: Where to Go from Here](chapters/ch20-where-to-go-from-here.md)