ReviseAlgo Logo

Type Hints & Annotations

mypy

Static type checking with mypy — configuring, running, understanding errors, and integrating type checking into your development workflow.

Interview: Professional code quality — shows commitment to reliable, maintainable codebases.

Last Updated: June 12, 2026 • 7 min read

mypy is the standard static type checker for Python. It analyzes your code without running it, catching type errors before they reach production. Combined with IDE integration, mypy catches bugs that would otherwise slip through testing.

Getting Started

  • Install: pip install mypy
  • Run: mypy mymodule.py or mypy --strict mypackage/
  • Configuration: mypy.ini or [mypy] section in pyproject.toml
  • IDE integration: most editors show mypy errors inline

Strictness Levels

  • Default: Only checks annotated code, ignores unannotated functions
  • --disallow-untyped-defs — require all functions to have annotations
  • --strict — all checks enabled, most rigorous
  • Use # type: ignore sparingly to suppress false positives

Common mypy Errors

  • Incompatible return type — function returns wrong type
  • Argument N to "func" has incompatible type — wrong argument type
  • Item "None" of "Optional[X]" has no attribute — forgot to check for None
  • Missing return statement — not all code paths return a value

Interview Insight

Know how to run mypy and interpret its errors. In interviews, mention that mypy catches type errors statically (without running code) and that --strict mode enforces full type coverage.

Use Cases

CI/CD pipeline — run mypy before merging pull requests

Legacy code migration — gradually add types and check with mypy

Library development — py.typed marker for typed library distribution

Bug prevention — catching type errors before runtime

Code review — mypy errors are objective, not subjective style preferences

Common Mistakes

Not running mypy at all — type hints without checking are just documentation

Overusing type: ignore — fix the root cause instead of suppressing errors

Starting with --strict on a legacy codebase — use gradual typing instead

Not configuring mypy.ini — default settings miss many useful checks

Forgetting py.typed marker — library consumers won't see your type hints