Effective Machine Learning Teams

Chapter 10. Building Blocks of Effective ML Teams

All happy families are alike; each unhappy family is unhappy in its own way.

Leo Tolstoy, Anna Karenina

The Anna Karenina principle contends that many factors must be in alignment for success, and that issues with any one of those many factors could lead to failure. The same may be said of effective ML teams, which must contend with all the complexities of teamwork in organizations, in addition to the complexities of ML product development.

But this principle needn’t doom our endeavors. On the contrary, as every environment in which teams operate—market, business, organizational structure, technology, data—is unique and dynamic, it stands that there is no cookie-cutter approach to success, and that every effective team must discover, and rediscover, its own unique path to success. In this chapter, we share some of the many tools we have at our disposal to move us closer to success, step by step.

As the Dalai Lama said: “Happiness is not something ready-made, it comes from your own actions.” So, it is crucial for individuals, teams, and whole organizations to experiment, reflect and adapt, and to have leadership guidance and support to do so. The building blocks of effective teams we present here help us work toward those ends at the level of individuals, teams, and organizations. In fact, we can think about the building blocks like the Swiss cheese model (introduced in Chapter 5): The more blocks we have in place, the less likely ...