In this chapter, we have looked at the benefits of product thinking in a custom project development environment. We studied why reusability matters and how we can build and integrate reusable software components at each stage of the data science project. We also went over the topic of finding the right balance between research and implementation. Finally, we looked at strategies for improving the reusability of our projects and explored the conditions that allow us to build standalone products based on our experience.
In the next section of this book, we will look at how we can build a development infrastructure and choose a technology stack that will ease the development and delivery of data science projects. We will start by looking at ModelOps, which is a set of practices for automating model delivery pipelines.
