This Post Has 6 Comments

1. Great Guide for GenAI on Google Cloud
   I’ve been slowly getting into LLMs and AI. My company uses GCP so this book was something I really needed to have. It was a great getting started guide that went in a pace that didn’t feel too slow or fast. I highly recommend for you to pick up this book if you’re looking to GenAI on GCP.

   Reply

2. Great Introduction to AI Internals!
   Have you used ChatGPT and OpenAI recently? Have you interacted with a ChatBot on a website when seeking assistance or support? Have you ever generated Text to Image, summarized a large group of documents or written a program using generative AI?Have you ever wondered what’s under the hood of these technologies? If so, then look no further than the book, Generative AI on Google Cloud with LangChain. Generative Artificial Intelligence (Gen AI) is now widely used worldwide and is expected to permeate all industries at some capacity.Generative AI on Google Cloud with LangChain is an excellent resource to reference if you desire an understanding of this technology. Author’s Lonid Kuligin, Jorge Zaldivar and Maximilian Tschochoei provides clear introductions of each topic and remain consistent for all thirteen chapters by providing a structured approach in defining terms and technical requirements, illustrating topics with real examples, summarizing and providing references for each chapter’s topics. There is an Appendix in the back of the book that goes into a greater detailed introduction of the topics covered in the first two chapters of this book and it would probably serve the reader well to read Appendix I before getting started to understand foundational models.To make the most of this book, a basic understanding of machine language (ML)and a basic understanding of Python allows you the engineer, scientist or enthusiast to follow along with each example.The reader should also setup a basic account with Google Cloud and setup their Google Cloud environment to use Vertex AI architecture platform by which the Google Cloud will interface with your Python scripts..Google Cloud is a public cloud of which the applications used in this book are built to interface with pre-trained foundational modelsaccessible via API.Large Language Models commonly referred to as LLMs encompassing a collection of tuned datasets with billions of parameters provide the frame of reference by which Python can interface with LLM-augmented autonomous agents to solve complex tasks. LangChain is derived from Language Chain and consists of building blocks called Runnables and Chains. The first chapter begins with a detailed explanation of how runnables and chains fit hand in hand so that basic definitions and formulas can be defined and passed to a “runnable” which is a predefined method or function that can be invoked in the Python script. Runnables can subsequently be executed within a chain logic of statements to produce in a structured or unstructured format of the response.The chain logic allows responses to be passed from one chain to the next in order to generate the appropriate response.Prompt Engineering is a “cornerstone” of generative AI which describes the art of crafting prompts, text and multimodal inputs that determine the outputs of pre-trained foundation models. Breakthroughs in pattern recognition, reasoning capabilities and predictive behaviour of human intent are the direct result of testing, tuning and further developing the logic which is described in great detail of this book.Retrieval Augmented Generation (RAG), “grounds responses” or centers the code to eliminate or reduce anomalies which can be described as hallucinations. If you’ve used Generative AI as an early adopter, you may be familiar with a hallucinated response which is a flaw in the answer, image or response of the queried model. This book covers in detail how your application can reduce occurrences of hallucinations programmatically.Generative AI on Google Cloud with LangChain also dives deeper into mathematics and discusses ranking responses or retrievals and utilizing rankings to vectorize data logic in stores to provide more accuracy. Tuning and testing LLMs and test scenarios are covered along with summarizing responses. An interesting concept of “closed book” vs “open book” answering system for MultiQueryRetriever results discusses relying strictly on the model’s predefined information versus importing external documents, images or source documents to generate a more informed response suitable to the query. There are a myriad of topics and appropriate Python scripts to cover each example and test on your own. The final chapters describe in greater detail how to build a complete application and property test and deploy your AI project. This book assures the reader that although generative AI has evolved tremendously; there is still human interaction that will be required for continuous development, testing, model tuning and API integrations. As I delve deeper into generative AI this book is an excellent reference and guide to understand the complexities and I will continue to keep it as a reference. Great read overall!

   Reply

3. Excellent Resource!
   Since this book was written by the creator of the topic LangChain it is an excellent reference.However it is not for a novice. As stated in the book someone with a strong foundation with engineering, developer, etc. Knowledge of python and machine learning is beneficial.You are provided free coding examples/files and step by step instructions. However you are required to have: Software/hardware covered in the book Operating system requirements Python with the required libraries installed Windows, macOS, or Linux and ChromeOS Vertex AI API on Google Cloud A custom Google Search enabled and a Google Cloud API key Vertex Vector Search, Vertex Agent Builder LangSmith (free tier) We encourage you to create a Python virtual environment (using Python venv or conda) and install all the dependencies in your virtual environment with pip or conda install.The book is very well written and walks you through different scenarios and processes. Generative AI, Creating Chatbots, etc. Provides references to other knowledge sources.I would definitely recommend the book to a seasoned IT professional wanting to learn about the topic.

   Reply

4. Great Book for Generative AI for Cloud
   The book is good for beginners as there are tons of resources online and this book tries to bring it all up in one place. I am new to cloud and generative AI so this book will personally help me to know more about it and make it efficient to use the knowledge in the book in my real-world projects.

   Reply

5. Comprehensive guide for building scalable GenAI apps with LangChain & VertexAI
   This book is well suited for developers, architects, and AI enthusiasts seeking to harness state-of-the-art GenAI technologies in a cloud environment. Building blocks are clearly explained using code snippets using Python-based implementations showing how to use LangChain and Vertex AI effectively.Provided insights into optimizing costs and performance when deploying on Google Cloud. Made readers to dive with practical examples like chatbot creation, document summarization, and image generation.Key topics covered from – Overview of GenAI concepts and applications, Introduction to LangChain for constructing AI (VertexAI on Google Cloud) workflows, Designing from data ingestion to model serving, Best practices.Readers will be able to develop GenAI solutions using LangChain with Google Cloud VertexAI and deploy AI models. Able to design optimized and effective AI pipelines by applying best practices for ethical and responsible AI implementation.Thanks again to publications for giving me this opportunity to review this book.

   Reply

6. Concise, hands on learning LangChain on Google Cloud
   After the GenAI , we are entering the AgenticAI era, this books will serve us not to get alone. This book introduces us into the foundations of GenAI applications using LangChain framework. After that, we learn RAG, in a practical point of view, always using LangChain in a easy way. We learn about Agents , Chatbots and Agentic workflows, and how to implement using Google Cloud tools in conjunction with LangChain. I found quite useful the chapters about multimodality, advanced techniques for parsing and ingesting documents, and specially how to work with long context. All in all, a great book to learn how to create GenAI applications with LangChain.

   Reply