Appendix: Deep Dives
These 15 deep-dive topics go beyond the tutorial chapters. Each topic provides foundational knowledge that helps you understand why Claude Code works the way it does — not just how to use it.
How to Use This Appendix
- During the tutorial: chapters link to relevant appendix topics when a concept deserves deeper exploration
- As reference: each topic is self-contained and can be read independently
- For research: topics include references to Anthropic's published research and the broader AI/ML literature
Topics by Category
Foundations
| # | Topic | Related Chapters | Depth |
|---|---|---|---|
| A01 | Prompt Engineering for Claude Code | Ch1-Ch6 | How to write effective prompts in a tool-calling context |
| A02 | Context Engineering | Ch3, Ch15 | The 2025-2026 paradigm shift from prompt engineering |
| A03 | LLM Fundamentals | Ch1, Ch3 | Transformer architecture, attention, tokenization |
Architecture
| # | Topic | Related Chapters | Depth |
|---|---|---|---|
| A04 | Agent Architecture Patterns | Ch8, Ch9, Ch15 | ReAct, plan-and-execute, multi-agent coordination |
| A05 | Tool Calling Internals | Ch1, Ch5 | How Claude selects tools, JSON Schema matching |
| A06 | MCP Protocol Deep Dive | Ch7 | Transport protocols, capability negotiation, production servers |
Economics & Optimization
| # | Topic | Related Chapters | Depth |
|---|---|---|---|
| A07 | Token Economics | Ch3, Ch15 | BPE tokenization, cost models, budget optimization |
| A10 | Prompt Caching | Ch3, Ch13 | Cache mechanics, prefix matching, cost impact |
| A11 | Model Selection Guide | Ch1, Ch15 | Opus vs Sonnet vs Haiku — when to use each |
AI Research
| # | Topic | Related Chapters | Depth |
|---|---|---|---|
| A08 | Constitutional AI & CLAUDE.md | Ch2 | How CLAUDE.md implements constitutional AI at project level |
| A09 | Extended Thinking | Ch8, Ch13 | When thinking helps, budget management, tool interaction |
| A12 | Safety & Alignment | Ch10 | Permission system as safety layer, responsible agent design |
Advanced Patterns
| # | Topic | Related Chapters | Depth |
|---|---|---|---|
| A13 | Multi-Agent Coordination | Ch9, Ch14 | Consensus, conflict resolution, scaling patterns |
| A14 | RAG vs Long Context | Ch7 | When retrieval beats long context, hybrid strategies |
| A15 | Claude Code Internals | Ch3, Ch14 | The harness loop, session management, compaction |
Reading Order Suggestions
If you're a developer new to AI: A03 (LLM Fundamentals) → A01 (Prompt Engineering) → A05 (Tool Calling) → A07 (Token Economics)
If you're building agent systems: A04 (Agent Architecture) → A13 (Multi-Agent) → A09 (Extended Thinking) → A15 (Claude Code Internals)
If you're optimizing costs: A07 (Token Economics) → A10 (Prompt Caching) → A11 (Model Selection) → A02 (Context Engineering)
If you're a team lead: A08 (Constitutional AI) → A12 (Safety) → A02 (Context Engineering) → A06 (MCP Protocol)