Devlin Nexley (204 results)

Condition: New.

Paperback. Condition: new. Paperback. Autonomous AI systems are moving beyond generating text and recommendations into environments where they can plan tasks, invoke tools, access enterprise data, modify systems, and coordinate with other agents. This shift creates a fundamentally different security problem: organizations must secure not only AI models and applications, but the autonomous execution layer surrounding them.This book presents a security engineering approach to designing, deploying, governing, and operating agentic AI systems in production environments. It examines how identity, authorization, runtime policy enforcement, Zero Trust principles, threat modeling, observability, and containment must evolve when software can make decisions and initiate actions dynamically.The architecture moves from the agent runtime and its trust boundaries through the major attack surfaces created by autonomous execution. It examines prompt injection, indirect instruction attacks, memory poisoning, retrieval manipulation, tool exploitation, credential abuse, data leakage, agent impersonation, and threats emerging from multi-agent collaboration. Rather than treating these as isolated vulnerabilities, the book analyzes how individual weaknesses can combine into larger attack paths across agent workflows.Particular attention is given to secure agent identity and delegated authorization, capability-based access control, policy-as-code, tool gateways, sandboxed execution, memory isolation, retrieval security, secrets management, and continuous verification. The discussion also examines how Zero Trust architecture can be adapted for autonomous agents by separating identity, intent, authorization, execution, and trust decisions.The book then moves into operational security. Agent telemetry, behavioral analytics, security operations, decision-path monitoring, incident response, execution freezing, identity quarantine, context rollback, memory recovery, and progressive privilege restoration are examined as components of an operational model for autonomous systems. These mechanisms are considered alongside enterprise governance, auditability, accountability, risk management, and compliance requirements.Designed for security architects, AI engineers, platform engineers, DevSecOps teams, enterprise architects, security leaders, and technology executives, this book provides an architecture-first perspective on securing AI agents at enterprise scale.Topics include: - Agent runtime security and autonomous execution architectures- Threat modeling for autonomous AI systems- Agent identity, authentication, delegation, and authorization- Zero Trust architecture for AI agents- Prompt injection and indirect instruction defense- Memory security and knowledge poisoning- Secure tool and function execution- Retrieval-augmented generation security- Multi-agent trust and inter-agent communication- Secrets and sensitive-data protection- Runtime policy enforcement and capability controls- Agent observability and behavioral security analytics- Security operations for autonomous AI environments- Incident response, containment, and recovery- Enterprise AI governance, auditability, and risk management- Resilient security architectures for production agentic platformsThe central engineering challenge is not simply preventing an AI agent from behaving incorrectly. It is designing an environment in which autonomous systems can operate with useful levels of capability while their identities, permissions, decisions, data access, and external actions remain continuously controlled and observable. Autonomous AI systems are moving beyond generating text and recommendations into env Shipping may be from multiple locations in the US or from the UK, depending on stock availability.…

Condition: New.

Condition: New.

Condition: New.

Condition: New.

Paperback. Condition: new. Paperback. Linux Mint provides a stable foundation for modern Linux desktop computing, but operating it effectively at a professional level requires more than knowing how to install applications or configure a desktop. Reliable workstation environments depend on sound architectural decisions, controlled system state, disciplined deployment practices, and a lifecycle strategy that accounts for change over time.This book takes a systems-engineering approach to Linux Mint, examining the operating system from the perspective of architecture, administration, deployment, automation, and operational reliability. Instead of presenting a collection of beginner tutorials, it explores how the major components of a Linux Mint workstation interact and how those components can be engineered into predictable and maintainable computing environments.The discussion moves from system architecture and initialization through storage, software delivery, hardware integration, desktop services, security boundaries, networking, automation, development workloads, virtualization, performance engineering, and enterprise deployment. Particular attention is given to the operational behavior of systems after deployment, where configuration drift, package dependencies, hardware differences, inconsistent policies, and uncontrolled upgrades can gradually undermine reliability.You will explore: - System architecture, initialization, service management, and desktop session lifecycle- Filesystem organization, storage design, partitioning considerations, and data persistence- APT repositories, dependency resolution, Flatpak, package governance, and software delivery- Cinnamon architecture, graphical sessions, desktop configuration, and user-environment state- Hardware detection, firmware, graphics drivers, peripheral integration, and compatibility considerations- Linux permissions, privilege boundaries, administrative controls, and workstation security architecture- NetworkManager, DNS resolution, connectivity, firewalling, and enterprise networking considerations- Shell environments, scheduled execution, scripting, and repeatable workstation automation- Developer workstation architecture for compilers, runtimes, source control, and isolated workloads- KVM/QEMU virtualization, containers, and practical workload-isolation models- CPU, memory, storage, boot, and service-level performance analysis- Corporate workstation deployment models and controlled endpoint provisioning- Imaging, cloning, identity regeneration, baseline construction, and deployment consistency- Configuration management, state reconciliation, drift detection, and policy enforcement- Upgrade rings, release transitions, recovery planning, technical debt, and long-term lifecycle managementThe central theme is operational control: how to make workstation environments reproducible without making them unnecessarily rigid, how to preserve flexibility without allowing uncontrolled divergence, and how to manage operating-system change without turning every upgrade into an emergency.Whether Linux Mint is being evaluated for professional workstations, development environments, managed desktop fleets, technical labs, or long-lived personal systems, the principles presented here provide a framework for understanding its behavior and engineering it for reliability.This is a technical reference for readers who want to move beyond desktop usage and understand the architectural and operational decisions that determine how Linux Mint systems behave in the real world. Linux Mint provides a stable foundation for modern Linux desktop computing, but operating it effectively at a professional level requires more than know Shipping may be from multiple locations in the US or from the UK, depending on stock availability.…

Condition: New.

Condition: New.

Condition: New.

Condition: New.

Paperback. Condition: new. Paperback. Modern software systems are no longer defined by application code alone. They are defined by the ability to store, process, replicate, coordinate, and analyze enormous volumes of data across distributed infrastructure while maintaining reliability, consistency, and performance under continuous change.As organizations move toward real-time analytics, event-driven architectures, cloud-native platforms, AI-powered applications, and globally distributed services, the complexity of managing data at scale has become one of the most challenging disciplines in modern engineering. Building these systems requires a deep understanding of storage engine internals, distributed coordination, consistency models, fault tolerance, stream processing, observability, and infrastructure behavior far beyond traditional application development.This book provides a comprehensive exploration of the architectures, principles, and implementation patterns that power today's largest and most demanding platforms. Rather than focusing on individual tools or vendor-specific technologies, it examines the underlying engineering concepts that govern how large-scale systems behave in production environments.Readers will explore the internals of distributed storage systems, transaction processing, replication architectures, partitioning strategies, consensus algorithms, event-driven systems, streaming platforms, analytical engines, and cloud-native infrastructure. The book explains not only how these systems are designed, but also how they behave under scale, contention, latency, network failures, and operational stress.The discussion extends into modern data infrastructure, including vector retrieval systems, feature stores, real-time machine learning pipelines, lakehouse architectures, and GPU-aware distributed platforms. Throughout the book, emphasis is placed on runtime behavior, performance bottlenecks, engineering tradeoffs, and the operational realities that emerge in production environments.Inside, you will learn how to: Design scalable and fault-tolerant distributed architecturesUnderstand storage engine internals, indexing structures, and persistence mechanismsImplement replication, partitioning, and consistency strategies for large-scale systemsAnalyze transactions, concurrency control, and distributed coordination protocolsBuild event-driven and stream-processing platforms capable of real-time computationEngineer resilient systems that withstand failures, outages, and unpredictable workloadsOptimize memory, storage, networking, and infrastructure utilizationDeploy and operate stateful distributed workloads in cloud-native environmentsImplement observability, tracing, monitoring, and reliability engineering practicesSecure distributed data platforms through modern governance and zero-trust principlesArchitect AI-native data infrastructure, vector retrieval systems, and machine learning pipelinesDiagnose production bottlenecks and understand failure patterns observed in real-world systemsDrawing from distributed systems engineering, database internals, cloud infrastructure, platform operations, and large-scale production architectures, this book presents a systems-focused approach to designing and operating modern data platforms. Whether building high-throughput transaction systems, real-time analytics pipelines, globally distributed services, search platforms, recommendation engines, or AI-native applications, the principles covered here provide a foundation for understanding how complex systems behave and evolve at scale. Drawing from distributed systems engineering, database internals, cloud infrastructure operations, and large-scale production architectures, th Shipping may be from multiple locations in the US or from the UK, depending on stock availability. …

Paperback. Condition: new. Paperback. Modern large language models are no longer confined to centralized cloud APIs. Advances in open-weight models, quantization, GPU acceleration, and local inference runtimes have made it possible to design AI systems that execute directly on developer workstations, private infrastructure, edge devices, and enterprise environments.LLaMA and Ollama in Production AI Systems examines the architecture behind this local AI stack, connecting the foundation-model layer represented by LLaMA with the runtime and inference layer provided by Ollama. Rather than treating local AI as a collection of command-line tools, the book approaches it as a systems-engineering problem involving model execution, memory, compute, APIs, deployment, observability, security, and operational trade-offs.The book explores how LLaMA-class transformer models are represented and executed, how Ollama manages local model execution, and how quantization and model packaging affect memory consumption, throughput, latency, and output quality. It examines the inference pipeline from tokenization and context processing through KV-cache management, sampling, streaming generation, and hardware acceleration.It also moves beyond single-machine experimentation into application and infrastructure architecture. You will examine local inference APIs, model routing, retrieval-augmented generation, tool-using applications, agentic workflows, hybrid local-cloud architectures, and deployment patterns for environments where privacy, latency, cost, availability, or data sovereignty influence where inference should occur.The book also addresses the operational realities that emerge when LLM inference becomes part of a production system: resource contention, model lifecycle management, observability, performance measurement, security boundaries, failure handling, capacity planning, and multi-node deployment.What You Will ExploreLLaMA model architecture and its role in the local LLM ecosystemOllama's runtime and model-execution architectureThe lifecycle of a local inference requestModel packaging, configuration, and runtime parametersGGUF and quantized model executionMemory requirements, KV caches, context windows, and token generationCPU, GPU, and hybrid inference trade-offsOllama's API layer and application integration patternsModel routing and multi-model local inferenceRetrieval-augmented generation with locally hosted modelsTool-using and agentic AI architecturesLocal-first application designHybrid local and cloud inference architecturesObservability and performance engineering for LLM workloadsSecurity, isolation, and data-governance considerationsEdge, on-premises, workstation, and enterprise deployment patternsCapacity planning and operational strategies for production systemsWho This Book Is ForThis book is intended for software engineers, AI engineers, ML engineers, platform engineers, DevOps and infrastructure professionals, solution architects, and technical leaders who want to understand how local LLM systems are engineered beyond basic model execution.It is particularly suited to readers who need to evaluate or build private AI infrastructure, integrate local inference into existing applications, design hybrid AI platforms, or understand the engineering consequences of moving LLM inference from centralized APIs to infrastructure they control.The goal is not simply to show how to run a model locally. It is to explain how the model, runtime, hardware, application, and infrastructure layers fit together-and how those layers behave when local LLM inference becomes a production engineering concern. Advances in open-weight models, Shipping may be from multiple locations in the US or from the UK, depending on stock availability.…

Condition: New.

Condition: New.

Condition: New.

Condition: New.

Condition: New.

Condition: New.

Condition: New.

Condition: New.

Condition: New.

Paperback or Softback. Condition: New. Microsoft 365 Copilot Systems Design. Book.

Paperback. Condition: new. Paperback. Linux Mint provides a stable foundation for modern Linux desktop computing, but operating it effectively at a professional level requires more than knowing how to install applications or configure a desktop. Reliable workstation environments depend on sound architectural decisions, controlled system state, disciplined deployment practices, and a lifecycle strategy that accounts for change over time.This book takes a systems-engineering approach to Linux Mint, examining the operating system from the perspective of architecture, administration, deployment, automation, and operational reliability. Instead of presenting a collection of beginner tutorials, it explores how the major components of a Linux Mint workstation interact and how those components can be engineered into predictable and maintainable computing environments.The discussion moves from system architecture and initialization through storage, software delivery, hardware integration, desktop services, security boundaries, networking, automation, development workloads, virtualization, performance engineering, and enterprise deployment. Particular attention is given to the operational behavior of systems after deployment, where configuration drift, package dependencies, hardware differences, inconsistent policies, and uncontrolled upgrades can gradually undermine reliability.You will explore: - System architecture, initialization, service management, and desktop session lifecycle- Filesystem organization, storage design, partitioning considerations, and data persistence- APT repositories, dependency resolution, Flatpak, package governance, and software delivery- Cinnamon architecture, graphical sessions, desktop configuration, and user-environment state- Hardware detection, firmware, graphics drivers, peripheral integration, and compatibility considerations- Linux permissions, privilege boundaries, administrative controls, and workstation security architecture- NetworkManager, DNS resolution, connectivity, firewalling, and enterprise networking considerations- Shell environments, scheduled execution, scripting, and repeatable workstation automation- Developer workstation architecture for compilers, runtimes, source control, and isolated workloads- KVM/QEMU virtualization, containers, and practical workload-isolation models- CPU, memory, storage, boot, and service-level performance analysis- Corporate workstation deployment models and controlled endpoint provisioning- Imaging, cloning, identity regeneration, baseline construction, and deployment consistency- Configuration management, state reconciliation, drift detection, and policy enforcement- Upgrade rings, release transitions, recovery planning, technical debt, and long-term lifecycle managementThe central theme is operational control: how to make workstation environments reproducible without making them unnecessarily rigid, how to preserve flexibility without allowing uncontrolled divergence, and how to manage operating-system change without turning every upgrade into an emergency.Whether Linux Mint is being evaluated for professional workstations, development environments, managed desktop fleets, technical labs, or long-lived personal systems, the principles presented here provide a framework for understanding its behavior and engineering it for reliability.This is a technical reference for readers who want to move beyond desktop usage and understand the architectural and operational decisions that determine how Linux Mint systems behave in the real world. Linux Mint provides a stable foundation for modern Linux desktop computing, but operating it effectively at a professional level requires m Shipping may be from our Sydney, NSW warehouse or from our UK or US warehouse, depending on stock availability.…

Paperback. Condition: new. Paperback. Modern software systems are no longer defined by application code alone. They are defined by the ability to store, process, replicate, coordinate, and analyze enormous volumes of data across distributed infrastructure while maintaining reliability, consistency, and performance under continuous change.As organizations move toward real-time analytics, event-driven architectures, cloud-native platforms, AI-powered applications, and globally distributed services, the complexity of managing data at scale has become one of the most challenging disciplines in modern engineering. Building these systems requires a deep understanding of storage engine internals, distributed coordination, consistency models, fault tolerance, stream processing, observability, and infrastructure behavior far beyond traditional application development.This book provides a comprehensive exploration of the architectures, principles, and implementation patterns that power today's largest and most demanding platforms. Rather than focusing on individual tools or vendor-specific technologies, it examines the underlying engineering concepts that govern how large-scale systems behave in production environments.Readers will explore the internals of distributed storage systems, transaction processing, replication architectures, partitioning strategies, consensus algorithms, event-driven systems, streaming platforms, analytical engines, and cloud-native infrastructure. The book explains not only how these systems are designed, but also how they behave under scale, contention, latency, network failures, and operational stress.The discussion extends into modern data infrastructure, including vector retrieval systems, feature stores, real-time machine learning pipelines, lakehouse architectures, and GPU-aware distributed platforms. Throughout the book, emphasis is placed on runtime behavior, performance bottlenecks, engineering tradeoffs, and the operational realities that emerge in production environments.Inside, you will learn how to: Design scalable and fault-tolerant distributed architecturesUnderstand storage engine internals, indexing structures, and persistence mechanismsImplement replication, partitioning, and consistency strategies for large-scale systemsAnalyze transactions, concurrency control, and distributed coordination protocolsBuild event-driven and stream-processing platforms capable of real-time computationEngineer resilient systems that withstand failures, outages, and unpredictable workloadsOptimize memory, storage, networking, and infrastructure utilizationDeploy and operate stateful distributed workloads in cloud-native environmentsImplement observability, tracing, monitoring, and reliability engineering practicesSecure distributed data platforms through modern governance and zero-trust principlesArchitect AI-native data infrastructure, vector retrieval systems, and machine learning pipelinesDiagnose production bottlenecks and understand failure patterns observed in real-world systemsDrawing from distributed systems engineering, database internals, cloud infrastructure, platform operations, and large-scale production architectures, this book presents a systems-focused approach to designing and operating modern data platforms. Whether building high-throughput transaction systems, real-time analytics pipelines, globally distributed services, search platforms, recommendation engines, or AI-native applications, the principles covered here provide a foundation for understanding how complex systems behave and evolve at scale. Drawing from distributed systems engineering, database internals, cloud infrastructure operations, and large-scale production arch Shipping may be from our Sydney, NSW warehouse or from our UK or US warehouse, depending on stock availability. …

PAP. Condition: New. New Book. Shipped from UK. THIS BOOK IS PRINTED ON DEMAND. Established seller since 2000.

Paperback. Condition: new. Paperback. Linux Mint provides a stable foundation for modern Linux desktop computing, but operating it effectively at a professional level requires more than knowing how to install applications or configure a desktop. Reliable workstation environments depend on sound architectural decisions, controlled system state, disciplined deployment practices, and a lifecycle strategy that accounts for change over time.This book takes a systems-engineering approach to Linux Mint, examining the operating system from the perspective of architecture, administration, deployment, automation, and operational reliability. Instead of presenting a collection of beginner tutorials, it explores how the major components of a Linux Mint workstation interact and how those components can be engineered into predictable and maintainable computing environments.The discussion moves from system architecture and initialization through storage, software delivery, hardware integration, desktop services, security boundaries, networking, automation, development workloads, virtualization, performance engineering, and enterprise deployment. Particular attention is given to the operational behavior of systems after deployment, where configuration drift, package dependencies, hardware differences, inconsistent policies, and uncontrolled upgrades can gradually undermine reliability.You will explore: - System architecture, initialization, service management, and desktop session lifecycle- Filesystem organization, storage design, partitioning considerations, and data persistence- APT repositories, dependency resolution, Flatpak, package governance, and software delivery- Cinnamon architecture, graphical sessions, desktop configuration, and user-environment state- Hardware detection, firmware, graphics drivers, peripheral integration, and compatibility considerations- Linux permissions, privilege boundaries, administrative controls, and workstation security architecture- NetworkManager, DNS resolution, connectivity, firewalling, and enterprise networking considerations- Shell environments, scheduled execution, scripting, and repeatable workstation automation- Developer workstation architecture for compilers, runtimes, source control, and isolated workloads- KVM/QEMU virtualization, containers, and practical workload-isolation models- CPU, memory, storage, boot, and service-level performance analysis- Corporate workstation deployment models and controlled endpoint provisioning- Imaging, cloning, identity regeneration, baseline construction, and deployment consistency- Configuration management, state reconciliation, drift detection, and policy enforcement- Upgrade rings, release transitions, recovery planning, technical debt, and long-term lifecycle managementThe central theme is operational control: how to make workstation environments reproducible without making them unnecessarily rigid, how to preserve flexibility without allowing uncontrolled divergence, and how to manage operating-system change without turning every upgrade into an emergency.Whether Linux Mint is being evaluated for professional workstations, development environments, managed desktop fleets, technical labs, or long-lived personal systems, the principles presented here provide a framework for understanding its behavior and engineering it for reliability.This is a technical reference for readers who want to move beyond desktop usage and understand the architectural and operational decisions that determine how Linux Mint systems behave in the real world. Linux Mint provides a stable foundation for modern Linux desktop computing, but operating it effectively at a professional level requires m Shipping may be from our UK warehouse or from our Australian or US warehouses, depending on stock availability.…

Paperback. Condition: new. Paperback. Modern large language models are no longer confined to centralized cloud APIs. Advances in open-weight models, quantization, GPU acceleration, and local inference runtimes have made it possible to design AI systems that execute directly on developer workstations, private infrastructure, edge devices, and enterprise environments.LLaMA and Ollama in Production AI Systems examines the architecture behind this local AI stack, connecting the foundation-model layer represented by LLaMA with the runtime and inference layer provided by Ollama. Rather than treating local AI as a collection of command-line tools, the book approaches it as a systems-engineering problem involving model execution, memory, compute, APIs, deployment, observability, security, and operational trade-offs.The book explores how LLaMA-class transformer models are represented and executed, how Ollama manages local model execution, and how quantization and model packaging affect memory consumption, throughput, latency, and output quality. It examines the inference pipeline from tokenization and context processing through KV-cache management, sampling, streaming generation, and hardware acceleration.It also moves beyond single-machine experimentation into application and infrastructure architecture. You will examine local inference APIs, model routing, retrieval-augmented generation, tool-using applications, agentic workflows, hybrid local-cloud architectures, and deployment patterns for environments where privacy, latency, cost, availability, or data sovereignty influence where inference should occur.The book also addresses the operational realities that emerge when LLM inference becomes part of a production system: resource contention, model lifecycle management, observability, performance measurement, security boundaries, failure handling, capacity planning, and multi-node deployment.What You Will ExploreLLaMA model architecture and its role in the local LLM ecosystemOllama's runtime and model-execution architectureThe lifecycle of a local inference requestModel packaging, configuration, and runtime parametersGGUF and quantized model executionMemory requirements, KV caches, context windows, and token generationCPU, GPU, and hybrid inference trade-offsOllama's API layer and application integration patternsModel routing and multi-model local inferenceRetrieval-augmented generation with locally hosted modelsTool-using and agentic AI architecturesLocal-first application designHybrid local and cloud inference architecturesObservability and performance engineering for LLM workloadsSecurity, isolation, and data-governance considerationsEdge, on-premises, workstation, and enterprise deployment patternsCapacity planning and operational strategies for production systemsWho This Book Is ForThis book is intended for software engineers, AI engineers, ML engineers, platform engineers, DevOps and infrastructure professionals, solution architects, and technical leaders who want to understand how local LLM systems are engineered beyond basic model execution.It is particularly suited to readers who need to evaluate or build private AI infrastructure, integrate local inference into existing applications, design hybrid AI platforms, or understand the engineering consequences of moving LLM inference from centralized APIs to infrastructure they control.The goal is not simply to show how to run a model locally. It is to explain how the model, runtime, hardware, application, and infrastructure layers fit together-and how those layers behave when local LLM inference becomes a production engineering concern. Advances in open-w Shipping may be from our Sydney, NSW warehouse or from our UK or US warehouse, depending on stock availability.…

Paperback. Condition: new. Paperback. Autonomous AI systems are moving beyond generating text and recommendations into environments where they can plan tasks, invoke tools, access enterprise data, modify systems, and coordinate with other agents. This shift creates a fundamentally different security problem: organizations must secure not only AI models and applications, but the autonomous execution layer surrounding them.This book presents a security engineering approach to designing, deploying, governing, and operating agentic AI systems in production environments. It examines how identity, authorization, runtime policy enforcement, Zero Trust principles, threat modeling, observability, and containment must evolve when software can make decisions and initiate actions dynamically.The architecture moves from the agent runtime and its trust boundaries through the major attack surfaces created by autonomous execution. It examines prompt injection, indirect instruction attacks, memory poisoning, retrieval manipulation, tool exploitation, credential abuse, data leakage, agent impersonation, and threats emerging from multi-agent collaboration. Rather than treating these as isolated vulnerabilities, the book analyzes how individual weaknesses can combine into larger attack paths across agent workflows.Particular attention is given to secure agent identity and delegated authorization, capability-based access control, policy-as-code, tool gateways, sandboxed execution, memory isolation, retrieval security, secrets management, and continuous verification. The discussion also examines how Zero Trust architecture can be adapted for autonomous agents by separating identity, intent, authorization, execution, and trust decisions.The book then moves into operational security. Agent telemetry, behavioral analytics, security operations, decision-path monitoring, incident response, execution freezing, identity quarantine, context rollback, memory recovery, and progressive privilege restoration are examined as components of an operational model for autonomous systems. These mechanisms are considered alongside enterprise governance, auditability, accountability, risk management, and compliance requirements.Designed for security architects, AI engineers, platform engineers, DevSecOps teams, enterprise architects, security leaders, and technology executives, this book provides an architecture-first perspective on securing AI agents at enterprise scale.Topics include: - Agent runtime security and autonomous execution architectures- Threat modeling for autonomous AI systems- Agent identity, authentication, delegation, and authorization- Zero Trust architecture for AI agents- Prompt injection and indirect instruction defense- Memory security and knowledge poisoning- Secure tool and function execution- Retrieval-augmented generation security- Multi-agent trust and inter-agent communication- Secrets and sensitive-data protection- Runtime policy enforcement and capability controls- Agent observability and behavioral security analytics- Security operations for autonomous AI environments- Incident response, containment, and recovery- Enterprise AI governance, auditability, and risk management- Resilient security architectures for production agentic platformsThe central engineering challenge is not simply preventing an AI agent from behaving incorrectly. It is designing an environment in which autonomous systems can operate with useful levels of capability while their identities, permissions, decisions, data access, and external actions remain continuously controlled and observable. Autonomous AI systems are moving beyond generating text and recommendat Shipping may be from our Sydney, NSW warehouse or from our UK or US warehouse, depending on stock availability.…

Paperback. Condition: new. Paperback. Autonomous AI systems are moving beyond generating text and recommendations into environments where they can plan tasks, invoke tools, access enterprise data, modify systems, and coordinate with other agents. This shift creates a fundamentally different security problem: organizations must secure not only AI models and applications, but the autonomous execution layer surrounding them.This book presents a security engineering approach to designing, deploying, governing, and operating agentic AI systems in production environments. It examines how identity, authorization, runtime policy enforcement, Zero Trust principles, threat modeling, observability, and containment must evolve when software can make decisions and initiate actions dynamically.The architecture moves from the agent runtime and its trust boundaries through the major attack surfaces created by autonomous execution. It examines prompt injection, indirect instruction attacks, memory poisoning, retrieval manipulation, tool exploitation, credential abuse, data leakage, agent impersonation, and threats emerging from multi-agent collaboration. Rather than treating these as isolated vulnerabilities, the book analyzes how individual weaknesses can combine into larger attack paths across agent workflows.Particular attention is given to secure agent identity and delegated authorization, capability-based access control, policy-as-code, tool gateways, sandboxed execution, memory isolation, retrieval security, secrets management, and continuous verification. The discussion also examines how Zero Trust architecture can be adapted for autonomous agents by separating identity, intent, authorization, execution, and trust decisions.The book then moves into operational security. Agent telemetry, behavioral analytics, security operations, decision-path monitoring, incident response, execution freezing, identity quarantine, context rollback, memory recovery, and progressive privilege restoration are examined as components of an operational model for autonomous systems. These mechanisms are considered alongside enterprise governance, auditability, accountability, risk management, and compliance requirements.Designed for security architects, AI engineers, platform engineers, DevSecOps teams, enterprise architects, security leaders, and technology executives, this book provides an architecture-first perspective on securing AI agents at enterprise scale.Topics include: - Agent runtime security and autonomous execution architectures- Threat modeling for autonomous AI systems- Agent identity, authentication, delegation, and authorization- Zero Trust architecture for AI agents- Prompt injection and indirect instruction defense- Memory security and knowledge poisoning- Secure tool and function execution- Retrieval-augmented generation security- Multi-agent trust and inter-agent communication- Secrets and sensitive-data protection- Runtime policy enforcement and capability controls- Agent observability and behavioral security analytics- Security operations for autonomous AI environments- Incident response, containment, and recovery- Enterprise AI governance, auditability, and risk management- Resilient security architectures for production agentic platformsThe central engineering challenge is not simply preventing an AI agent from behaving incorrectly. It is designing an environment in which autonomous systems can operate with useful levels of capability while their identities, permissions, decisions, data access, and external actions remain continuously controlled and observable. Autonomous AI systems are moving beyond generating text and recommendat Shipping may be from our UK warehouse or from our Australian or US warehouses, depending on stock availability.…