What this book covers
The book offers a structured narrative, starting with an introduction to generative AI and its integration with cloud computing. This is followed by an exploration of the model layer, diving deeper into the intricacies of Large Language Models (LLMs), including the evolution of Natural Language Processing (NLP) and the advent of transformer models. It discusses techniques such as fine-tuning and Retrieval-Augmented Generation (RAG) for augmenting model knowledge. The book then discusses prompt engineering methods. Moving on to the application level, it covers the development framework and strategies, emphasizing scaling, security, safety, and compliance with responsible AI principles. The concluding section provides foresight into the future trajectory of generative AI. Here is the outline of the chapters in this book:
Chapter 1, Cloud Computing Meets Generative AI: Bridging Infinite Impossibilities, introduces the concept of LLMs, what ChatGPT is based on, and their significance in conversational and generative AI. It examines the generative capabilities of LLMs, such as text generation and creative writing. The chapter concludes by exploring the practical applications of LLMs and their future directions in virtual assistants, content creation, and beyond.
Chapter 2, NLP Evolution and Transformers: Exploring NLPs and LLMs, takes you on a journey through the evolution of transformers – the heart of LLMS, from preceding technology known as Natural Language Processing (NLP) to how a powerful new paradigm has now been created using NLP and LLMs.
Chapter 3, Fine-Tuning: Building Domain-Specific LLM Applications, talks about the benefits of fine-tuning, different techniques of fine-tuning, how to align models to human values with RLHF, evaluating fine-tuned models, and real-life examples of fine-tuning success.
Chapter 4, RAGs to Riches: Elevating AI with External Data, discusses the fundaments of vector databases and how they play a critical role in building a Retrieval-Augmented Generation (RAG) based application. We will also explore chunking strategy evaluation techniques along with a real-life case study.
Chapter 5, Effective Prompt Engineering Strategies: Unlocking Wisdom Through AI, takes a look at prompt engineering with ChatGPT and some techniques to not only make prompts more effective but also understand some of the ethical dimensions of prompting.
Chapter 6, Developing and Operationalizing LLM-Based Cloud Applications: Exploring Dev Frameworks and LLMOps, uses a software application developer lens to focus on areas that would support developer activities such as programmatic application development frameworks, allowing for AI-enabled applications. We will also look at the lifecycle management of generative AI models in addition to operationalizing the management of generative AI models, along with exciting topics such as agents, autonomous agents, and assistant APIs.
Chapter 7, Deploying ChatGPT in the Cloud: Architecture Design and Scaling Strategies, explores how to scale a large deployment of a generative AI cloud solution. You’ll gain an understanding of limits, design patterns, and error handling while taking a look at areas and categories that ensure a large-scale generative AI application or service will be robust enough to handle a large number of prompts.
Chapter 8, Security and Privacy Considerations for Gen AI: Building Safe and Secure LLMs, uncovers existing and emerging security threats related to GenAI models, and how to mitigate them, by applying security controls or other techniques to ensure a safe, secure environment. We will also cover a concept known as red-teaming, as well as auditing and reporting.
Chapter 9, Responsible Development of AI Solutions: Building with Integrity and Care, delves into the essential components required to construct a secure generative AI solution, emphasizing the key principles of responsible AI and addressing the challenges of LLMs through these principles. It also explores the escalating concern over deepfakes, their harmful impacts on society, and strategies for developing applications with a responsible AI-first approach. Additionally, it examines the current global regulatory trends and the burgeoning start-up ecosystem in this domain.
Chapter 10, The Future of Generative AI: Trends and Emerging Use Cases, is one of the most exciting chapters in this book, discussing the future of generative AI solutions, highlighting hot emerging trends such as the rise of small language models, offering predictions, exploring the integration of LLMs on edge devices, and examining the impact of quantum computing and the path to AGI.
