How does the introduction of operator overloading in Python enhance its functionality for numerical computations?

Operator overloading in Python allows custom objects to interact with operators, such as +, -, *, and /, mirroring their use with built-in data types. This feature gains significance in numerical computations, particularly when working with complex mathematical types like matrices, vectors, or data structures in scientific libraries like NumPy. By enabling operator overloading, Python lets developers define intuitive operations on custom classes, which can improve code readability and maintainability. What are some examples of operator overloading enhancing Python's capabilities in numerical contexts, and how might it influence the performance and scalability of Python applications?