The Recommendation Engine for Shoes

In the nineties, Amazon's goal was to get big fast. The monolif and a massive data basin for structure in the back end couldn't grow any more from an architectural point of view. So they started carving off pieces of the monolif into smaller building blocks that each have their own unique scaling and reliability capabilities. And then we see the effectiveness of the teams tapering off as these small services work on the same thing.