type-safety

Type hints specify parameter and return types in function signatures. Add them using syntax like `def process(data: str) -> dict:` to enable static analysis tools like mypy to detect type mismatches before runtime, improving code reliability and IDE support.

Optional[T] is shorthand for Union[T, None] and explicitly marks a value that may be None. Use Optional for nullable fields and Union when a value can be one of several types. Both prevent NoneType errors by forcing explicit None checks.

TypeVar and Generic enable you to write reusable functions and classes that work with multiple types while preserving type safety. Define `T = TypeVar('T')` and use it in signatures like `def reverse(items: list[T]) -> list[T]` so mypy validates type consistency.

Strict mypy configuration enforces comprehensive type checking across all code paths, catching edge cases like implicit Optional values and unannotated parameters. It prevents type-related bugs from reaching production and makes refactoring safer.

Protocol defines structural types for duck typing, Literal restricts values to specific constants, and TypeAlias names complex types for clarity. These advanced features express precise type relationships that mypy validates, reducing runtime surprises in complex codebases.

Add type hints incrementally, starting with public APIs and data models. Use mypy's gradual typing to monitor progress without rewriting everything at once. Focus on high-risk functions where type errors cause bugs, building confidence before expanding coverage.