Books & Courses¶

Long-form resources sit here, separate from topic files, because they represent a different commitment class. You can scan a paper in 20 minutes; a book is weeks of work; a course is months. This page is for readers who want the deeper dive.

Books¶

AI Engineering — Chip Huyen, O'Reilly, 2025. The canonical textbook for building production LLM systems end-to-end: evaluation, finetuning, inference, agents, RAG, observability. Paywalled; companion repo chiphuyen/aie-book is open. [book]
Build a Large Language Model from Scratch — Sebastian Raschka, Manning, 2024. Implementation-first walkthrough of a small LLM: tokenization through attention through fine-tuning. Paywalled; companion repo rasbt/LLMs-from-scratch is free and the primary artifact for most readers. [book]
Hands-On Large Language Models — Jay Alammar and Maarten Grootendorst, O'Reilly, 2024. The practical counterpart to Raschka's first-principles book: applied patterns for text classification, search, clustering, and generation. Companion repo at HandsOnLLM/Hands-On-Large-Language-Models. [book]
Machine Learning Systems — Harvard CS249r, open textbook, 2024–ongoing. Not AI-security-specific, but the best freely available systems view of how ML pipelines actually run in production — necessary context for reasoning about supply-chain and training-time threats. [book]

Courses¶

Microsoft AI Red Team Training — Microsoft, 2025. Multi-module hands-on training covering PyRIT, adversarial testing methodology, and program-building for AI red teams. Free with a Microsoft Learn account. [course]
Anthropic's API and Agent-building Courses — Anthropic, ongoing. Free courses on prompt engineering, tool use, and real-world agent patterns. Primary authorship by the model provider. [course]
Microsoft AI Agents for Beginners — Microsoft, 2025. 11-lesson open course on agent fundamentals with runnable notebooks. Not security-focused, but a strong prerequisite for anyone who will red-team agentic systems. [course]