Thanks for the question, Nader. The short answer is that data engineers focus on infrastructure, while data scientists focus on analysis and predictions.
My blog post on the skill differences between data engineers and data scientists provides a lot more detail.
I believe in most cases it makes sense for bot makers to build their own natural language parser, rather than using a third party API. There are good strategic and technical arguments for doing this, and I want to show how easily you can put something together…
Unfortunately, TensorFlow only works on Mac (and Linux) right now. You can use it on a Windows using a virtual machine or using cloud computing tools, but for this tutorial, I decided not to discuss these topics.