Books

Move from observing AI agents to operating them through runtime evidence, evaluation, economics, reliability, and governed action.This practitioner-focused book develops an architecture for production AI systems where traces, evaluations, safety signals, cost, business outcomes, and operational authority become part of one evidence model.

A short, hands-on guide to reverse engineering with AI agents and REA 6.4.0: drive Ghidra from Claude Code, Codex or Cursor, check every claim against evidence, and work through native, Electron, Android, .NET and web targets you are allowed to analyse. Every output shown is a real lab run.

Curious about agentic AI but not sure where to start? This beginner-friendly guide takes you from the basics to building your own AI agents, even if you have no coding experience. Learn how agents work, choose the right tools and turn your ideas into practical, reliable solutions for real-world problems.

Kill an AI agent mid-refund and the retry bills every model call again and refunds twice. This book crashes one agent on purpose and recovers it on DBOS, Temporal, Restate and Prefect, with the real output of every run.

Learn how Prompt, Context, Harness, and Loop turn AI-generated code into reviewable engineering work. Visual explanations cover task framing, trustworthy context, safe tools, independent verification, and controlled iteration.

Learn the C++ fundamentals that connect source code, memory, ownership, build tools, and operating-system boundaries. Clear diagrams and small examples guide you toward reading, testing, and diagnosing systems-oriented programs.

Use AI-assisted engineering from core concepts to secure production workflows. Visual explanations and a continuing FleetOps case study connect context design, coding, testing, delivery, operations, and bounded agents with practical verification and human accountability.

Your LLM pipeline passed every check, and a fallback model still served 50 of its 71 successful calls. This case study follows one real multi-stage pipeline through its failures and shows how to make each stage traceable, attributable and honestly measured.

Decision models answer a typed question with a calibrated probability in one forward pass: no token stream, no parsing, no values outside your schema. This guide explains TypeSafe's Jev and the open "JEV-like" family, then shows how to choose, run, serve, fine-tune and evaluate local variants on your own hardware.

Polars 2.0 changed defaults that don't raise errors: lazy queries now stream, and joins no longer keep row order. This short guide shows exactly what breaks when you upgrade from 1.x, with every example run under both 1.44.2 and 2.0.0 and the real output printed side by side.

We will define the developer contract first, and we will then build the platform underneath it together, step by step, using decisions that hold up in regulated environments and in small startups alike. Along the way, we will provision clusters, wire delivery, publish a developer API, enforce policy and ship observability, and we will do all of it where you can see every command, every manifest and every failure. Nothing here stays theoretical, because platform problems only show their true shape where an abstraction meets a real cluster.

The battle-tested prompt framework and architectural system rules to turn any AI coding assistant into a principal full-stack engineer.

Você viu um modelo de linguagem fazer meia hora de trabalho sozinho. Este livro trata da pergunta que veio logo depois: até que ponto você pode confiar nele?

Has visto a un modelo de lenguaje hacer media hora de trabajo por su cuenta. Este libro trata de la pregunta que vino después: ¿hasta qué punto puedes confiar en él?

Skip the trial and error. 53 agent skills picked for Claude Code and Codex, with setup for both, each with who it fits, install commands, and a first run.