How is it any different in any way than any other async function? You just end up providing an unusually large number of parameters via headers and body, then at some point in the future the request completes with values and/or errors. What separates them?