This book bridges the gap between theoretical machine learning (ML) and its practical application in industry. It serves as a handbook for shipping production-grade ML systems, addressing challenges often overlooked in academic texts. Drawing on their experience at several major corporations and startups, the authors focus on real-world scenarios, guiding practitioners through the ML lifecycle, from planning and data management to model deployment and optimization. They highlight common pitfalls and offer interview-based case studies from companies that illustrate diverse industrial applications and their unique challenges. Multiple pathways through the book allow readers to choose which stage of the ML development process to focus on, as well as the learning strategy ('crawl,' 'walk,' or 'run') that best suits the needs of their project or team.
Rezensionen / Stimmen
'I love when practitioners share their hard-earned wisdom. This book doesn't shy away from the messy realities of data work, from sourcing to compliance. The case studies are especially valuable, showing how their framework holds up in real-world use cases.' Chip Huyen, author of AI Engineering and Designing Machine Learning Systems 'This book by Mohamed El-Geish, Shabaz Patel, and Anand Sampat is an invaluable reference for engineers and managers building best-in-class ML and AI systems. It provides practical guidance on essential considerations, methods, and tools, enabling teams to confidently navigate the complexities of real-world AI development and deployment.' Hassan Sawaf, aiXplain 'Shipping Machine Learning Systems is the rare book that goes beyond algorithms to show what it really takes to build production ML systems. It combines clear explanations with honest discussions of trade-offs at every stage, grounded in real examples from industry leaders like Instacart and WhatsApp. An essential guide for anyone serious about shipping robust ML products.' Riham Selim, Meta 'There is a significant difference between developing a machine learning system in a controlled lab environment and deploying it in production to serve real users. This book bridges that critical gap with clarity and depth. It is an invaluable resource for machine learning practitioners and application developers seeking to bring cutting-edge ML systems into the real world - reliably, safely, and at scale.' Emad Elwany, AI Technology Executive 'Shipping machine learning systems is where theory meets the real world, and this book delivers the practical guidance every engineer needs to succeed. It covers the unglamorous but essential work of deploying, monitoring, and scaling models in production. Having built AI systems at Kolena, I found the lessons here refreshingly real and immediately useful. This is the book I would hand any team building serious ML products.' Mohamed Elgendy, Kolena
Sprache
Verlagsort
Produkt-Hinweis
Illustrationen
Worked examples or Exercises
Gewicht
ISBN-13
978-1-009-12420-1 (9781009124201)
Copyright in bibliographic data and cover images is held by Nielsen Book Services Limited or by the publishers or by their respective licensors: all rights reserved.
Schweitzer Klassifikation
Mohamed El-Geish is CTO and Co-Founder of Monta AI. He has built machine learning systems used daily by millions worldwide. He led Amazon's Alexa Speaker Recognition and Cisco's Contact Center AI, co-founded Voicea (acquired by Cisco), contributed to products at LinkedIn and Microsoft, and co-authored 'Computing with Data' (2019). Shabaz Patel is Associate Director of Applied AI at Best Buy, where he architects scalable ML systems powering search and discovery experiences for millions of users. Previously, at One Concern, he spearheaded innovations in AI-driven climate risk mitigation. Educated at Stanford and IIT, he specializes in scalable MLOps and impactful AI deployments and founded Datmo, an ML startup. Anand Sampat is CTO and Co-Founder of OpsPro AI. He is an ML Leader and serial entrepreneur. He previously co-founded Datmo (acquired by One Concern) and led ML Solutions for One Concern, led ML for New Products at PathAI, and led ML at SambaNova Systems.
Autor*in
Monta AI
Best Buy
OpsPro AI
Co-Autor*in
Bank of America
Preface; Introduction; Part I. Ready, Aim, Fire, Aim, Fire, ...: 1. Planning; 2. Data; 3. Model development; 4. Model deployment and beyond; 5. Compute optimizations; Part II. Case Studies: 6. Nauto: data and model management; 7. Kavak: ML serverless architecture for car sales; 8. Instacart: journey in building Griffin; 9. WhatsApp: enhancing ML operations for fraud and abuse detection model; 10. ShortlyAI: Your AI writing partner; References; Index.
