language Introduced in L3

Python

The default language of data, ML, and quick scripts. Increasingly the second language even in JS-first stacks.

Mindmap

hover · click to navigate
this tech depends on / used by alternative Shipyard anchor
What it is

The plain-English version

Python is a general-purpose, high-level programming language. Released 1991. Reads almost like pseudocode. The default for data science, scripting, and machine learning. Web backend role is real but smaller than Node's.

Why it exists

The problem it solves

Python's gravity is in the data/ML world: NumPy, pandas, PyTorch, TensorFlow, scikit-learn — the libraries that defined modern ML. For backends, FastAPI and Django are credible choices. AI agents handle Python well; for any data pipeline, expect Python.

What it competes with

Alternatives

AlternativeTypeWhen it wins
TypeScriptlanguageJavaScript with a type system. The language most production codebases use, even ones that look like JavaScript.
Node.jsruntimeJavaScript outside the browser. The runtime that powers most modern web backends, build tooling, and AI agent CLIs.
Where it shows up in Shipyard

Deep links

Vocabulary

The words you'll hear

pip
The package installer. pip install requests.
venv / virtualenv
Isolated Python environments per project.
uv, poetry
Modern dependency tooling. uv is the new fast standard.
requirements.txt / pyproject.toml
Old / new manifests.
PEP 8
The style guide. black formats to it automatically.
List comprehension
[x*2 for x in nums] — Python's signature shape.
GIL
Global Interpreter Lock. Why CPython doesn't truly thread. Use multiprocessing or async.
Prompting

Bad vs. good prompt for Python

✕ Bad prompt
write a python script
✓ Good prompt
Write a Python 3.11 script that reads a CSV (path as CLI arg with argparse), validates each row with pydantic v2 against a Task model, prints a summary of valid/invalid rows. Use type hints throughout. Keep it under 80 lines. Show the pyproject.toml dependencies block too.

Why it works: Specifies Python version, validation library version (pydantic v1 vs v2 has different APIs), CLI tool, type hints, size budget, and asks for the dependency declaration. Concrete enough to ship.

Pitfalls

What bites real teams

⚠ Python 2 examples

Stack Overflow still has Python 2 answers. print as a statement is a giveaway. Always specify Python 3.x.

⚠ Pydantic v1 vs v2

Major API change. AI agents mix them. Pin and specify.

⚠ Async vs sync confusion

Python's asyncio is its own world. Mixing sync and async libraries deadlocks in subtle ways. Pick a side per app.

References

Official docs only