Google Machine Learning and Generative AI for Solutions Architects: ?Build efficient and scalable AI/ML solutions on Google Cloud

by Kieran Kavanagh, Priyanka Vergadia (Foreword by)

Overview

Architect and run real-world AI/ML solutions at scale on Google Cloud, and discover best practices to address common industry challenges effectively

• Understand key concepts, from fundamentals through to complex topics, via a methodical approach

• Build real-world end-to-end MLOps solutions and generative AI applications on Google Cloud

• Get your hands on a code repository with over 20 hands-on projects for all stages of the ML model development lifecycle

• Purchase of the print or Kindle book includes a free PDF eBook

Nearly all companies nowadays either already use or are trying to incorporate AI/ML into their businesses. While AI/ML research is undoubtedly complex, the building and running of apps that utilize AI/ML effectively is tougher. This book shows you exactly how to design and run AI/ML workloads successfully using years of experience some of the world’s leading tech companies have to offer. You’ll begin by gaining a clear understanding of essential fundamental AI/ML concepts, before moving on to grasp complex topics with the help of examples and hands-on activities. This will help you eventually explore advanced, cutting-edge AI/ML applications that address real-world use cases in today’s market. As you advance, you’ll recognize the common challenges that companies face when implementing AI/ML workloads, and discover industry-proven best practices to overcome these challenges. The chapters also teach you about the vast AI/ML landscape on Google Cloud and how to implement all the steps needed in a typical AI/ML project. You’ll use services such as BigQuery to prepare data; Vertex AI to train, deploy, monitor, and scale models in production; as well as MLOps to automate the entire process. By the end of this book, you will be able to unlock the full potential of Google Cloud's AI/ML offerings.

• Build solutions with open-source offerings on Google Cloud, such as TensorFlow, PyTorch, and Spark

• Source, understand, and prepare data for ML workloads

• Build, train, and deploy ML models on Google Cloud

• Create an effective MLOps strategy and implement MLOps workloads on Google Cloud

• Discover common challenges in typical AI/ML projects and get solutions from experts

• Explore vector databases and their importance in Generative AI applications

• Uncover new Gen AI patterns such as Retrieval Augmented Generation (RAG), agents, and agentic workflows

This book is for aspiring solutions architects looking to design and implement AI/ML solutions on Google Cloud. Although this book is suitable for both beginners and experienced practitioners, basic knowledge of Python and ML concepts is required. The book focuses on how AI/ML is used in the real world on Google Cloud. It briefly covers the basics at the beginning to establish a baseline for you, but it does not go into depth on the underlying mathematical concepts that are readily available in academic material.

Product Details

ISBN-13:	9781803247021
Publisher:	Packt Publishing
Publication date:	06/28/2024
Sold by:	Barnes & Noble
Format:	eBook
Pages:	552
File size:	15 MB
Note:	This product may take a few minutes to download.

About the Author

Kieran Kavanagh is a Principal Architect at Google Cloud, working with Google's largest retail customers and driving some of the industry's most challenging digital transformation and generative AI initiatives. Before joining Google, he was a Principal AI/ML Solutions Architect in Strategic Accounts at Amazon Web Services (AWS), building some of the most complex AI/ML systems in the world. He was also a Principal Architect at AT&T, leading their Mobile Internet infrastructure design, and he is a public speaker on the topics of AI/ML, MLOps, and large-scale cloud transformation. Originally from Cork, Ireland, he now lives in Atlanta, GA, with his wife, Katelyn.

Table of Contents

Table of Contents

1. AI/ML Concepts, Real-World Applications, and Challenges
2. Understanding the ML Model Development Lifecycle
3. AI/ML Tooling and the Google Cloud AI/ML Landscape
4. Utilizing Google Cloud's High-Level AI Services
5. Building Custom ML Models on Google Cloud
6. Diving Deeper—Preparing and Processing Data for AI/ML Workloads on Google Cloud
7. Feature Engineering and Dimensionality Reduction
8. Hyperparameters and Optimization
9. Neural Networks and Deep Learning
10. Deploying, Monitoring, and Scaling in Production
11. Machine Learning Engineering and MLOps with GCP
12. Bias, Explainability, Fairness, and Lineage
13. ML Governance and the Google Cloud Architecture Framework
14. Advanced Use Cases and Technologies
15. An Introduction to Generative AI
16. Generative AI on Google Cloud
17. Advanced Generative AI Concepts and Use Cases
18. Bringing It All Together—Building ML Solutions with GCP and Vertex