Effective Machine Learning Teams

Chapter 2. Product and Delivery Practices for ML Teams

Product development isn’t easy. In fact, most product development efforts fail, and the most common reason for failure is building the wrong product.

Henrik Kniberg, agile and Lean coach

You can practice shooting [basketballs] eight hours a day, but if your technique is wrong, then all you become is very good at shooting the wrong way. Get the fundamentals down and the level of everything you do will rise.

Michael Jordan

In Chapter 1, we introduced the five disciplines that are required for delivering ML solutions: product, delivery, ML, software engineering, and data. Later on, in Part II of the book, we’ll focus on many engineering, ML, and data practices to help teams build the thing right and reduce toil, waste, and rework. These practices will improve velocity and product quality. However, it’s important that we first start with product and delivery practices that help teams with an even more important goal: how to build the right thing.

In this chapter, we’ll focus on aspects of the ML product delivery lifecycle where we often see teams’ effort go to waste due to a lack of clarity or misalignment between what the customers or the business need and what the product engineering team delivers. We’ll introduce product and delivery practices that have helped us in our real-world ML projects. This chapter is organized to address three key phases of product delivery:

Discovery

To help teams understand and define the opportunity ...