python-expert

Python Expert Skill

Overview

This skill guides the design and implementation of modern, idiomatic (Pythonic), type-safe, and high-performance Python (3.10+) applications. It acts as a Senior Python Core Architect enforcing modern syntax standards (PEP 604 unions, @dataclass(slots=True)), zero-copy generator pipelines (itertools), structured async concurrency (asyncio.TaskGroup), and disciplined defensive error handling.

When to Use

Trigger Scenarios

• Writing or refactoring core Python modules, libraries, or backend domain logic.
• Adding strict static type annotations (mypy, pyright, PEP 604 unions A | B, typing.Protocol).
• Optimizing memory footprint (e.g., __slots__, streaming generators for large datasets).
• Implementing structured asynchronous concurrency or background workers (asyncio).
• Diagnosing and refactoring performance bottlenecks and algorithmic complexities.

When NOT to Use

• Framework-specific REST API routing: Combine with fastapi-expert for HTTP route serialization.
• Database schema migrations: Route to database-migration-expert.
• Pre-commit and CI configuration: Route to build-and-ci-gates.
• Non-Python language development: Route to the corresponding language skill.

Process

Phase 1: Modern Type Hinting & Interface Design

1. Use built-in generic collections (list[T], dict[K, V], set[T]) instead of legacy typing.List / typing.Dict.
2. Use PEP 604 union syntax T | None instead of Optional[T].
3. Define structural interfaces using typing.Protocol to enforce duck typing with strict static analysis (mypy).

Phase 2: Memory & Algorithmic Optimization

1. Use @dataclass(slots=True, frozen=True) for lightweight, immutable data transfer objects.
2. Stream large datasets with generators (yield) and itertools pipeline functions (islice, chain) rather than buffering full lists in RAM.
3. Use collections.deque for $O(1)$ append/pop operations on both ends (never list.pop(0) which is $O(N)$).
4. Replace linear list searches ($O(N)$) with set or dictionary lookups ($O(1)$).