What happens between a working prototype and a production release? This four-chapter ebook was created with the answer in mind, offering a focused look at system design, data preparation, pipeline orchestration, and deployment. Learn to uncover design weaknesses before implementation, explore faster data access and processing, and follow a batch inference example built with Kubeflow Pipelines. Then examine how to test inference under realistic conditions and compare strategies for releasing model updates. You’ll get concrete examples of the engineering decisions beyond model development as well as practical guidance for your next machine learning project.
about the authors
Valerii Babushkin is an accomplished data science leader with extensive experience. He currently serves as a Senior Principal at BP.
Arseny Kravchenko is a seasoned ML engineer currently working as a Senior Staff Machine Learning Engineer at Instrumental.
At Netflix,
Ville Tuulos designed and built Metaflow, a full-stack framework for data science. Currently, he is the CEO of a startup focusing on data science infrastructure.
Benjamin Tan Wei Hao leads a team of ML engineers and data scientists at DKatalis.
Shanoop Padmanabhan is a software engineering manager at Continental Automotive.
Varun Mallya is a senior ML engineer at DKatalis.
Chi Wang is a principal software developer in the Salesforce Einstein group.
Donald Szeto was the co-founder and CTO of PredictionIO.