Appendices

The best reference material is the kind you keep reaching for long after you have finished the book.
A Pragmatic Educator

Part Overview

The appendices provide essential reference material organized into five groups. Foundations and Setup covers mathematical prerequisites, ML essentials, Python tooling, environment configuration, and version control. Reference Materials includes the glossary, hardware guides, model cards, prompt templates, and benchmark catalogs. Framework Guides offers hands-on introductions to HuggingFace, LangChain, LangGraph, CrewAI, LlamaIndex, Semantic Kernel, and DSPy. Infrastructure and MLOps covers experiment tracking, inference serving, distributed computing, and containerization. Finally, Ecosystem Overview maps the broader LLM tooling landscape.

Appendices: 22 (Appendix A through V) covering reference material, framework tutorials, and infrastructure guides.

Big Picture

These appendices serve two purposes: they provide prerequisite refreshers you can consult before diving into specific chapters, and they offer practical framework guides you will return to when building real projects. Think of them as the book's toolbox.

Foundations and Setup

Appendix A Mathematical Foundations

Linear algebra, probability, calculus, and information theory essentials for understanding LLM internals.

Appendix B Machine Learning Essentials

Learning paradigms, loss functions, optimization, regularization, and evaluation metrics.

Appendix C Python Libraries and Patterns for LLM Development

Essential libraries, virtual environments, Jupyter/Colab setup, and common scripting patterns.

Appendix D Development Environment Setup

Step-by-step guides for configuring your local development environment and cloud workspaces.

Appendix E Git, DVC, and Reproducibility

Version control for code, data, and experiments with Git, DVC, and reproducibility best practices.

Reference Materials

Appendix F Glossary

Definitions of key terms, acronyms, and concepts used throughout the book.

Appendix G GPU Hardware and Cloud Compute

GPU selection, cloud provider comparison, and cost optimization strategies.

Appendix H Model Cards and Selection Guide

Quick-reference cards for major LLMs with selection criteria and comparison tables.

Appendix I Prompt Template Catalog

Ready-to-use prompt templates for common tasks: summarization, classification, extraction, and more.

Appendix J Datasets, Benchmarks, and Leaderboards

Curated list of datasets, evaluation benchmarks, and leaderboard resources for LLM research.

Framework Guides

Appendix K HuggingFace: Transformers, Datasets, and Hub

Practical guide to the HuggingFace ecosystem: loading models, fine-tuning with Trainer, and sharing on the Hub.

Appendix L LangChain: Chains, Agents, and Retrieval

Building LLM applications with LangChain's chain composition, agent framework, and retrieval integration.

Appendix M LangGraph: Stateful Agent Workflows

Graph-based agent orchestration with persistent state, branching, and human-in-the-loop patterns.

Appendix N CrewAI: Multi-Agent Orchestration

Multi-agent collaboration with role-based agents, tasks, and crew orchestration.

Appendix O LlamaIndex: Data Indexing and Query Engines

Data connectors, indexing strategies, and query engines for building RAG applications.

Appendix P Semantic Kernel: Enterprise AI Orchestration

Microsoft's SDK for integrating LLMs into enterprise applications with plugins and planners.

Appendix Q DSPy: Programmatic Prompt Optimization

Declarative prompt programming with automatic optimization and modular pipeline composition.

Infrastructure and MLOps

Appendix R Experiment Tracking: W&B and MLflow

Logging experiments, comparing runs, and managing model artifacts with Weights & Biases and MLflow.

Appendix S Inference Serving: vLLM, TGI, and SGLang

High-throughput LLM serving with continuous batching, PagedAttention, and production deployment patterns.

Appendix T Distributed ML: PySpark, Databricks, and Ray

Scaling data pipelines and training across clusters with PySpark, Databricks, and Ray.

Appendix U Docker and Containers for LLM Deployment

Containerizing LLM applications with Docker, GPU passthrough, and orchestration with Docker Compose.

Ecosystem Overview

Appendix V LLM Tooling Ecosystem

A survey of the broader LLM tooling landscape: evaluation platforms, guardrail libraries, observability tools, and emerging frameworks.

What Comes Next

Return to the Table of Contents to explore specific chapters, or visit the Front Matter for reading pathways tailored to your background.