Callbacks β Promises β async/await
One thread, never waiting
Node runs your code on a single thread. That sounds like a weakness until you see the trick: when your code asks for something slow β a database row, an HTTP call, a file β Node hands the wait to the operating system and moves on to the next request. When the answer arrives, Node comes back and runs the code that was waiting on it. One thread, thousands of connections, because it is never sitting idle.
This is non-blocking I/O, and it is the whole reason Node is good at APIs. Your job as the author is simple to state and easy to break: never make that one thread sit and wait. Everything in this module is a variation on that one rule.
The senior pitch: "Node isn't fast because it's parallel β it's fast because it refuses to wait. One thread stays busy by never blocking on I/O."