10-book AI engineering reading list for 2026, ordered from ML basics to production LLM systems
mdancho84 · x · 2026-11-23
Data scientist mdancho84 published an ordered 10-book reading list for leveling up in AI engineering, arguing it beats 1,000 hours of YouTube tutorials.
- AI Literacy Fundamentals (Ben Jones): hallucinations, cost structures, model limits.
- Software Engineering for Data Scientists (Catherine Nelson): testing, Git, Docker, logging — bridging notebook to production code.
- The StatQuest Illustrated Guides (Josh Starmer): intuition for how models learn and how transformers work.
- Build a Large Language Model From Scratch (Sebastian Raschka): tokenization, embeddings, context windows, and the mechanics of hallucination.
- AI Engineering (Chip Huyen): his top pick — prompting, RAG, fine-tuning, evals, security, model selection.
- Prompt Engineering for Generative AI (Phoenix & Taylor): direction, format, examples, evaluation, plus RAG pipelines and agent patterns.
- Generative AI System Design Interview (Aminian & Sheng): end-to-end architecture for chatbots, legal RAG, Cursor-style assistants.
- Co-Intelligence (Ethan Mollick): human-in-the-loop rollout and the just-me / delegated / automated framework.
- The Alignment Problem (Brian Christian): reward hacking, via the boat-racing AI that spins in circles farming power-ups.
Suggested entry points: coders start at book 4, ML practitioners at book 5, builders at book 6 working backward when stuck.
More from coding & agent
- Data Scientist's pick: 'Software Engineering for Data Scientists' bridges notebooks to production code — mdancho84 · 2026-11-23
- 10 books beat 1,000 hours of YouTube: a data scientist's ordered AI engineering reading path — mdancho84 · 2026-11-23
- TinyFish CPO opens his GitHub to show the AI reviewer and ops agent shipping to production — aakashgupta · 2026-11-23
- Pinokio launches Agent Messenger: chat with all your local apps like friends — cocktailpeanut · 2026-11-23
- Windows 11 caught running hidden telemetry that tracks which AI coding agents you use — cyb3rops · 2026-11-23
- Anthropic: skip RAG for knowledge bases under 200K tokens, cut costs up to 90% with prompt caching — CatAstro_Piyush · 2026-11-23