Large language models

Build smarter AI systems that go beyond the limits of large language models. Retrieval-Augmented Generation is a practical guide to designing, implementing, and scaling RAG applications with modern retrieval techniques, vector databases, and real-world deployment strategies.

Every test you have ever written assumed the same input gives the same output. Large language models broke that assumption, and with it most of what our profession knows about verification. This is the discipline that replaces it — golden datasets, property assertions, LLM judges you can actually trust, retrieval evaluation, adversarial suites, and CI gates that let a pipeline say no to a model.

Break free from expensive cloud APIs and reclaim your digital sovereignty by running powerful LLMs locally. Master the professional stack—from vLLM serving and Unsloth fine-tuning to advanced quantization on consumer GPUs. Build secure, lightning-fast AI applications that protect your data and eliminate recurring token fees forever. Stop renting your intelligence and start building a private, autonomous future on your own hardware.

Il giudizio non si delega. Il perimetro entro cui l'agente si muove, la verifica che deve passare, la condizione che lo ferma: quelli sì. Come un Senior trasforma vent'anni di mestiere nei vincoli che governano un agente, e come decisioni, criteri e lezioni restano quando la sessione finisce: è l'amplificazione del titolo.

AI is most dangerous when it's almost right. Fifteen essays on software engineering in the age of AI, read through the psychology of Alfred Adler, about the machine that is almost right, and about our almost-right responses to it.

What if agentic AI was less about hype and more about work you can actually inspect?OpenClaw and Hermes for Agentic AI is a practical field guide to building with local agents, memory, tools, sessions, cron jobs, and ACP bridges. It shows how to turn fuzzy ideas into reliable workflows with clear boundaries, real artifacts, and human review where it matters most.If you want a grounded look at how agent systems really operate, this book starts there.

Your agent passed every load test and timed out in production anyway. This free practitioner's guide shows why the traditional performance playbook breaks for agentic AI, and what replaces it: latency budgets, bounded autonomy, token SLOs, and production-grade observability, from a performance engineering and observability architect with two decades in production.

Reliable Generative AI bridges business use and technical architecture. It teaches the foundations of prompt design, RAG, agentic workflows, tool use, structured outputs, safety patterns, and evaluation without assuming the reader is a software engineer. The focus is practical: understanding how AI workflows fail, how to design around those failures, and how to build systems that professionals can trust.

Una guida pratica a Claude Code, scritta con Claude Code.Su Claude Code circolano due tipi di contenuti: video YouTube che si fermano ai primi dieci minuti e thread social che promettono "guide complete" in cambio di un'email. Questo libro nasce dall'esigenza opposta: avere finalmente un manuale che spieghi lo strumento per intero, senza hype e senza muri di lead generation.140 pagine costruite a partire dalla documentazione ufficiale Anthropic, verificate su Claude Code v2.1.123, con esempi reali invece che screenshot da marketing. Si parte dall'installazione su macOS, Linux e Windows, si passa per Plan Mode, CLAUDE.md, gestione del contesto, sicurezza dei permessi, Skill personalizzate, subagent custom, integrazione MCP, e si arriva al prompt engineering aggiornato al 2026 — quello vero, dopo che le formule magiche del 2023 si sono sgonfiate.Il libro è pensato per due lettori in parallelo: lo sviluppatore senior che vuole integrare Claude Code nel proprio workflow di produzione, e chi si avvicina per la prima volta agli strumenti AI agentici e ha bisogno di una mappa. I capitoli iniziali costruiscono le basi senza dare nulla per scontato; quelli avanzati entrano nei meccanismi che fanno la differenza tra un uso superficiale e uno strategico.Tre cose che troverai e che altrove fatico a trovare:Il prompt engineering riletto al 2026. Niente ruoli posticci, niente "agisci come un esperto di...". Le tre leve che contano davvero — istruzioni esplicite, contesto adeguato, esempi curati — applicate al contesto specifico della CLI.Workflow reali, non demo da slide. Esempi concreti su WordPress e Node.js presi da progetti veri, non da repository giocattolo.In apertura c'è una prefazione scritta da Claude Opus, a cui ho dato il manoscritto completo chiedendogli un commento onesto, anche dove il libro non funzionava. È rimasta lì come l'ha scritta: è il modo più trasparente che mi sia venuto in mente per inquadrare il libro.Il libro è gratis — pay-what-you-want con minimo zero. Se ti torna utile e vuoi contribuire, decidi tu quanto. Se non puoi o non vuoi, scarichi e leggi senza pensieri: la licenza Creative Commons BY-SA 4.0 ti permette anche di redistribuirlo e modificarlo, basta citare la fonte e mantenere la stessa licenza.

ဥပမာအားဖြင့်၊ သင်က AI ကို "မိုးရွာရင် ထီးယူသွားပါ" ဟု ခိုင်းထားလျှင်၊ မိုးသည်းထန်စွာ ရွာနေသော်လည်း အိမ်ခေါင်မိုး ပြိုကျနေပါက ၎င်းသည် ထီးကိုသာ ကိုင်၍ အိမ်ထဲတွင် ငုတ်တုတ်ထိုင်နေပေလိမ့်မည်။ အဘယ်ကြောင့်ဆိုသော် "အိမ်ခေါင်မိုး ပြိုလျှင် ပြေးပါ" ဟူသော စည်းမျဉ်းကို သင်က ထည့်မပေးထား သောကြောင့် ဖြစ်သည်။

This book gives senior technology leaders a practical operating system for enterprise AI strategy. It turns scattered pilots into a board-ready plan by walking through strategy, economics, vendor decisions, ownership, governance, GenAI architecture, and roadmap assembly: AI Platform Scorecard, Use-Case Prioritization Matrix, Budget Model + Cost Guardrails, Vendor Decision Framework, RACI Matrix.

You leave Claude Code running and come back to something missing. This book is arranged by the accident, not the mechanism: databases dropped, files deleted, git history rewritten, the rewind command taking work with it. Three of the four get the exact hook that blocks the command before it runs, with the recovery if it already happened; no PreToolUse hook sees the rewind command, and the book gives the settings-level lever and names the one place a refusal could still live. Appendix B prints all 195 commands the guards were fired at, refusals and allowances both.

Code Is the Side Effect"Software engineers are not primarily code writers. We are clarity traders — and that hasn't changed."You've seen the demos. The AI builds a whole feature from a sentence. The agent writes tests, fixes the failing ones, opens the PR. It's remarkable.Then you come back three months later. The codebase is a tangle. Nobody knows why anything is the way it is. The agent that built it has no memory of what it decided or why. And every time you ask it to add something new, it breaks two things you didn't know were connected.This is the pattern that nobody talks about. AI coding tools make the easy parts of engineering dramatically easier. They leave the hard parts untouched — and they create new hard parts that didn't exist before.Ways of Working is the book for engineers who want to work with AI agents rather than be gradually replaced by them — who understand that the tools are genuinely powerful and genuinely limited, and want to build practices that get the most from each.What you will actually learnThe world model framework. Before an agent can build anything well, it needs to understand what it's building and why. This book teaches you to give agents what they need: a structured, queryable representation of your architecture, your component contracts, your behavior specifications, and your code patterns. No world model = no sustained agentic development.Intent documentation. The most expensive bug in agentic codebases is not a hallucination — it's a decision made without context. Why is this rule here? Why is this boundary where it is? Agents can't infer rationale from code. You have to write it down.Spec-Kit and formal specifications. GitHub's Spec-Kit brings machine-readable, traceable, CI-verified specifications to engineering teams. This book shows how to use it to turn requirements into agent inputs that are precise enough to generate correct implementations.Graph explainers. Tools like Graphify and Understand-Anything transform codebases and documents into queryable knowledge graphs — giving agents navigable context instead of flat text. This is the memory substrate that makes multi-agent systems reliable at scale.Agent architecture that holds. What makes an agent coherently itself? When do file-based agent systems break down and what replaces them? How does constraint-based coordination (borrowed from holocracy) solve the autonomy-coherence problem that has stumped AI researchers for decades?Claude Code, for real. A complete treatment of Claude Code's CLAUDE.md convention, permission model, hooks, and slash commands. Plus the oh-my-claudecode ecosystem: 15+ specialized agents, workflow orchestration patterns (autopilot, ralph, ultrawork), and the skills framework for team-specific automation.The AI-native organization. What genuine AI-native teams look like beneath the marketing. How to hire, structure, and lead them. What language-oriented programming and constrained natural language mean for the future of the human-code relationship.Who it's forEngineers who are past the "should I use AI?" question and into the "how do I use it without losing my engineering integrity?" question.Senior engineers. Engineering managers. Technical leaders. People who have noticed that the more they delegate to AI, the less certain they feel — and who want to understand why.From the AuthorI've been building production systems with AI agents for years. Not demos — systems that had to work reliably across months, maintain themselves as requirements changed, and produce outputs that engineers could understand and defend.That experience has made me skeptical in both directions.Skeptical of the "AI will do everything" vision — because I've watched too many AI-generated codebases collapse under the weight of accumulated misunderstanding.Equally skeptical of the "nothing fundamentally changed" position — because the engineers who treat AI coding tools as just faster autocomplete are making a category error they'll pay for in months of maintenance debt.Something genuinely new is happening. This book is my attempt to think about it clearly.

Your clients pay you for careful admin work. An AI agent can do the first draft of most of it overnight — if it knows exactly what "done" looks like and exactly what it must never touch.