Implementing MLOps in the Enterprise

Chapter 3. Getting Started with Your First MLOps Project

If you’re itching to get started with building your MLOps project and pipelines, you’re at the right chapter. Surprisingly (or not, if you’ve been carefully reading the book so far), the first step doesn’t require a notebook or IDE. Instead, it requires a proper discussion with the decision makers at your company. AI and ML open up new technological frontiers, but outside of academia, they have to be connected to a business use case. This is what makes them valuable to people. Therefore, the first thing to do is to figure out the business use case that justifies your project, as well as the goals and the expected ROI.

Once the business side is clear, it’s time to go to your computer. But don’t open your notebook just yet. The next step is to plan the ML project. This includes the resources you will need, processes that will run, prototyping the solution, the pipeline structure, and the design. Once these components are approved, it’s finally time to develop your ML pipeline. In this chapter, we explore each of these stages in detail.

Identifying the Business Use Case and Goals

AI is transforming businesses and global economies. A PwC report predicts that AI could contribute as much as $15.7 trillion to the global economy by 2030. Moreover, 45% of total economic gains will come from product enhancements, stimulating consumer demand. This is because AI will drive greater product variety, with increased personalization, attractiveness, ...