Annotations describe what a function expects. Python does not enforce them at runtime. Tools such as type checkers and editors read them, and so do people.

def lengths(
    words: list[str],
) -> dict[str, int]:
    return {w: len(w) for w in words}

• list[str], dict[str, int] and tuple[int, int] describe contents.
• int | None means either one.
• type UserId = int creates a named alias.

A type parameter links an input type to an output type:

def last[T](items: list[T]) -> T:
    return items[-1]

Now a checker knows that last(["a"]) is a str.

Since Python 3.14, annotations are evaluated only when something asks for them, so a class can name itself in its own methods without quotes. typing.get_type_hints(fn) returns the evaluated annotations.

Add exact annotations: word_lengths(words: list[str]) -> dict[str, int], and make first_or_none generic with a type parameter T, taking items: list[T] and returning T | None. Define the alias type Scores = dict[str, int]. Fix first_or_none so that it returns None for an empty list.

The explanation opens here once every check passes.