Books

"The model is safe." "The filter is effective." "We red-teamed it and found nothing." "Wereviewed a sample and it looked fine." Each of those is a claim about evidence, and each is unfalsifiable as stated. Reviewing thirtyoutputs and finding no problems is consistent with a failure rate of one in eleven. Threehundred tests against a thousand possible failure modes leave seven hundred nobody looked at. This book supplies the missing arithmetic — and it was written without generating a singleharmful output to study.

A model that works in development and fails in production usually fails for reasons that havenothing to do with the model. It waits behind other requests. It loads slowly after a restart.It answers differently because somebody moved a decoding parameter by a tenth of a point. Aguardrail blocks a third of legitimate traffic and reports itself as working. None of those appear in a benchmark. They appear in an incident — and almost every one of themis invisible in an instantaneous measurement and obvious in a trend.

Quantise the weights. Batch the requests. Add a draft model. Prune what you can. Shorten thecache. A long list of techniques and a short list of reasons. This book measures the reason first: the machine streams memory at 22.54 GB/s and computes at116.58 GFLOP/s, so anything performing fewer than 5.17 operations per byte is waiting. Decodingperforms 0.22. Every technique in the book is then measured against that one number — andseveral turn out to be addressing the wrong side of it.

Move Beyond Simple Prompts. Build Reliable, Autonomous AI Systems.Connecting an LLM to an API is easy; building an AI agent that works reliably in production is hard. Mastering AI Agents gives you the exact architectural frameworks, workflows, and engineering practices needed to design, deploy, and scale intelligent agents that solve real-world problems.Practical Engineering: Master memory systems, tool integration, multi-agent orchestration, and system security.Production-Ready Focus: Learn how to evaluate, monitor, and maintain agents so they remain controllable and safe.Evergreen Principles: Built on core software architecture principles that stay valuable even as frameworks and models evolve.Grab your copy today and start building the next generation of intelligent software!

Teaser TextExplore six advanced AI applications shaping the future.Build intelligent solutions for smart farming, AI-powered content creation, predictive healthcare, smart-city traffic optimization, conversational AI, and AI-generated music.From predicting crop yields and monitoring health to optimizing traffic and creating music with GANs, this volume demonstrates the practical possibilities of modern Artificial Intelligence.Six projects. Six domains. One journey toward advanced AI innovation.

Teaser TextBuild the future with advanced Artificial Intelligence.The final volume of the 50 AI Projects series presents 10 advanced projects covering AI code review, news bias detection, blockchain voting, deepfake detection, smart farming, AI content creation, predictive healthcare, smart-city traffic optimization, conversational AI, and AI-generated music.Explore emerging AI technologies while following the complete Software Development Life Cycle—from problem definition and system architecture to development, testing, deployment, and future enhancement.50 projects. 5 volumes. One complete journey into practical AI development.

Teaser TextEnter the world of advanced AI engineering.Volume 4 presents 10 advanced Artificial Intelligence projects covering medical diagnosis, crime prediction, autonomous drones, predictive maintenance, disaster management, conversational AI, financial fraud detection, legal document analysis, smart shopping, and autonomous vehicle vision.Go beyond basic models and explore how complex AI systems can be planned, designed, developed, tested, deployed, and enhanced using a complete Software Development Life Cycle.From intelligent models to intelligent systems.

Понятный иллюстрированный путеводитель по ключевым концепциям и практическим применениям больших языковых моделей. Идеально подходит для работы над проектами, подготовки к собеседованиям и самостоятельного обучения.

Teaser TextTake your AI projects to the next level.Explore 10 intermediate-level Artificial Intelligence projects covering NLP, Machine Learning, Computer Vision, prediction, classification, sentiment analysis, image recognition, and voice-based applications.From fake news detection and student performance prediction to resume ranking, plant disease identification, sign language recognition, and language translation, Vol-2 takes you deeper into practical AI development through a complete Software Development Life Cycle.Learn the concepts. Build the systems. Think like an AI developer.

Teaser TextLearn AI by building real projects.Explore 10 practical beginner-level Artificial Intelligence projects and learn how to take an idea from problem definition to design, coding, testing, implementation, documentation, and future enhancement.From chatbots and face detection to resume screening, NLP, emotion detection, and intelligent recommendation systems, Vol-1 provides a structured introduction to practical AI development through the complete Software Development Life Cycle.Don’t just study AI. Build it.

Prompt engineering got you a demo; context engineering ships it. This hands-on book teaches you to decide what goes into the model's context window — retrieval, memory, assembly, compression, caching — by building ctxkit, a real Python context engine, one chapter at a time.

Build the Next Generation of Enterprise AI SystemsCloud-Powered Agentic AI is a practical guide to designing, developing, deploying, and operating production-ready AI applications using Large Language Models (LLMs) and modern cloud-native technologies.From prompt engineering and Retrieval-Augmented Generation (RAG) to AI agents, multi-agent systems, Model Context Protocol (MCP), FastAPI, Docker, Kubernetes, and LLMOps, this book provides a complete roadmap for building scalable, secure, and maintainable AI solutions.Through hands-on examples, architecture diagrams, case studies, and a comprehensive capstone project, you will learn how to move beyond prototypes and engineer enterprise-grade AI systems.Inside this book, you'll learn how to:Build LLM-powered applications from scratchDesign Retrieval-Augmented Generation (RAG) pipelinesDevelop AI agents and multi-agent workflowsIntegrate tools using the Model Context Protocol (MCP)Deploy AI services with FastAPI, Docker, and KubernetesApply LLMOps for monitoring, evaluation, and continuous deliverySecure and govern AI applications in enterprise environmentsBuild a complete cloud-native AI platform from development to productionWhether you are an AI engineer, software developer, cloud architect, or machine learning practitioner, this book will equip you with the practical skills needed to build reliable and scalable AI systems in the era of generative AI.

This book is all about testing and closing the gap between infrastructure health and output quality, and how easily that gap can be filled for weeks by a failure that's gone unnoticed.This book looks at the less obvious side of the problem. In the next few chapters, we're going to run 41 experiments on one application. About half of them will give you a result that the system doesn't report. It's not a dig at anyone's engineering, since the systems in question are competently built and mostly behave well.

Running LLMs locally is easy. Running them well is an engineering problem. Practical LLM Inference gets into the details that matter, from quantization and GGUF internals to GPU offloading, benchmarking and production deployment. Learn how to size hardware, find bottlenecks and build fast, reliable inference systems without the AI hand-waving.