How is it like to be an API consumer and an API provider at the same time? How is building an orchestration layer for distributed data different than building an API backed by a single database? When building an app that goes across many domains, when is the time to sync data and when is it okay to proxy? How to deal with downstream failures and communicate that efficiently to our customers? The Netflix Studio often builds orchestration layers to tackle and abstract a lot of the complexities that are behind various micro services - let’s see how GraphQL fits perfectly into that story and helps making those layers a better experience for developers on both sides of the API.