This book explores how Artificial Intelligence (AI) and Software-Defined Networking (SDN) can transform the way modern networks are designed, secured, and operated. In an era shaped by cloud computing, IoT, 5G, and edge computing, traditional network management is no longer enough--this book reveals how AI agents bring autonomy, intelligence, and adaptability to meet these challenges.
The book starts with the foundational concepts in AI and SDN, guiding readers toward advanced architectures and real-world applications. It examines urgent needs such as scalable, self-healing networks and proactive cybersecurity, showing how AI techniques--including reinforcement learning, graph neural networks, and explainable AI--can achieve intent-based networking, cognitive healing, federated learning, and intelligent automation. Each chapter combines conceptual overviews with detailed discussions, case studies, and actionable insights, making it accessible to students, researchers, engineers, and decision-makers alike. It bridges technical depth with broader considerations such as ethics, governance, energy efficiency, and disaster recovery. It unifies AI and networking into a single, practical framework rather than treating them as separate fields. The inclusion of curated resources, from books and blogs to courses and glossaries, supports ongoing learning beyond the text itself.
This book serves as a roadmap, guiding readers in designing intelligent, secure, and adaptive network ecosystems--essential for those aiming to lead the next generation of decentralized, resilient, and AI-driven digital infrastructure.
What you will learn:
- Understand core AI-agent architectures and their integration with Software-Defined Networking for scalable, adaptive environments.
- How to use ML, deep learning, reinforcement learning, and graph neural networks to optimize, automate, and secure networks.
- How to develop AI-enabled networks with real-world case studies from telecom, smart cities, and enterprise IT.
- Explore trends like federated learning, edge AI, programmable optical networks, and AI-driven disaster recovery
Who this book is for:
This book serves network architects and engineers using AI-driven automation to solve scalability and complexity challenges. It guides AI researchers and data scientists applying advanced methods for smarter, more efficient networks. Security professionals will also find value in AI-driven threat detection, incident response, and collaborative defense.
About the Author :
Het Mehta is a seasoned software engineer with nearly seven years at Cisco, specializing in AI-driven networking and SD-WAN innovation. Currently serving as a Software Engineer IV in San Jose, he has led impactful projects enhancing network reliability, scalability, and security. Het redesigned serial list synchronization for network authentication, integrating REST APIs with Neo4j for seamless and secure controller updates. He played a key role in improving network stability by decoupling control and data planes, ensuring uninterrupted MPLS and private connections. His creation of vDiagnose, a proactive SD-WAN health assessment tool, has boosted uptime and operational efficiency across deployments. Het has also developed selective logging features that streamline troubleshooting, cut noise, and speed resolution times. Earlier roles saw him engineering highly scalable, distributed control plane overlays and optimizing kernel footprints for diverse devices. With deep expertise in distributed systems, diagnostics, and enterprise networking, Het combines technical precision with a strong focus on user experience. His work consistently bridges innovation and practical deployment, making modern networks more resilient, adaptive, and intelligent. He is based in the US.
