[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8},"actionData","errors","root",{"_28":-5,"_38":-5,"_39":40,"_41":42,"_43":44,"_45":46,"_47":48,"_49":50,"_51":52,"_53":54,"_55":56,"_57":58},"routes/_store+/_readers+/email_author.$slug/route",{"_9":10,"_11":-5},"book",{"_12":13,"_14":15,"_16":17,"_18":19,"_20":21,"_22":23},"currentUser","id","Z2lkOi8vbGVhbnB1Yi9TdG9yZTo6Qm9va1Byb2R1Y3QvNzkyMjk","title","The Agentic Engineer","state","published","contactAuthorMethod","email","aboutText","

AI/LLM books rot fast. This one bets on what doesn't. They teach the model that's hot this quarter, the SDK with the cleanest tutorial, the prompt that worked on a demo. Six versions later, none of it survives.

This book is the opposite bet. It teaches mental models, decision frameworks, and named failure patterns for systems that have large language models in load-bearing paths: the kind of knowledge that outlives any specific model, vendor, or framework.

It's written for working engineers and tech leads who are shipping agentic systems in production. Not researchers. Not prompt-engineering hobbyists. People who have to operate this stuff at 3am when it breaks, and need a vocabulary for what they're looking at.

The evidence base is real production war stories. A virtual employee that shipped a brief ending mid-sentence because the API returned `stop_reason: \"max_tokens\"` and nothing in the system raised. A prompt-cache flag the SDK accepted but the provider silently no-op'd. An MCP tool surface that lied about resource lifecycle. An agentic reviewer that swept the wrong directory and called the work done. Every named principle is anchored to an incident that actually happened in a shipped system.

What's inside:

A reading vocabulary for probabilistic-inside-deterministic systems
Context, tool surfaces, and eval loops as engineering primitives, not afterthoughts
A catalog of named failure modes you'll recognize the next time you see them
Orchestration patterns and observability disciplines for production agentic systems
A decision framework for when to use an agent vs. a script vs. a human
A structural playbook for building teams around agentic work

What's NOT in this book: model comparisons, framework tutorials, \"future of work\" speculation, prompt cookbooks, or any sentence that depends on which API version was current the week the book shipped. Those live in the book's online appendix, versioned and dated.

If you've ever debugged an LLM-in-the-loop system at 3am and felt that your normal engineering instincts were the wrong tools for the job, this book is for you.

","authors",[24],{"_12":25,"_26":27,"_28":29},"Z2lkOi8vbGVhbnB1Yi9TdG9yZTo6UHJvZHVjdEF1dGhvci84NDgwMw","name","Victor Velazquez","user",{"_30":31},"avatar",{"_32":33,"_34":35,"_36":37},"url","https://d39qdlcrvnra4b.cloudfront.net/avatars/3728582/medium/avatar.jpg?1779805804","styleName","medium","exists",true,"shoppingCartId","shoppingCart",{"_12":71,"_72":73,"_74":37,"_75":76,"_77":78,"_79":80,"_81":37},"locale","en","readerTheme","light","constants",{"_69":70},"publishers",{"_65":66,"_67":68},"learnerOrganizations",[],"needsLocaleCookieSync",false,"announcements",{"_59":60,"_61":-5},"sentryEnvironment","production","sentryRelease","cce21d19a0d9612afdbced29947de1d8868e588b","announcement_store",{"_62":52,"_63":64},"announcement_authors","flag","text","

Kick off your book project, get started with GhostAI, get better at marketing, or spend the day doing all three! Free live workshops on Zoom. Saturday, June 27, 2026.

\n","companies",[],"universities",[],"paypalClientId","AX53Cwz8jR6rVVzzZiZNWWXl0dGagi-GZrjG-S3Ua14E5nrKMn7DxDQ730O4R8LusLPp49ALn6aRVaEy","no-cart","count",0,"free","updatedAt","","paymentGateway","STRIPE","cartItems",[],"loading"]

Email the Author