Every term, defined 📖

argument

A value you pass into a function when you call it. Different from a parameter, which is the name the function gives it. Lesson 17

assert

A statement that raises AssertionError if its condition is false. The simplest form of a test. Lesson 29

async / await

Keywords for concurrency: async def makes a coroutine that can pause at await points, letting thousands of tasks share one thread while they wait. Lesson 40

boolean

A value that is either True or False. Named after George Boole. Lesson 6

bug

A gap between what you believe your code does and what it actually does. Found by making beliefs visible and testing them. Lesson 10

class

A blueprint that bundles data (attributes) and the operations on it (methods) into one named type. Lesson 31

closure

A function that remembers variables from the scope where it was defined, even after that scope has finished. The basis of decorators. Lesson 19

comprehension

A one-line way to build a list, dict or set from a loop, e.g. [n*n for n in range(5)]. Lesson 16

context manager

An object used with with that guarantees setup and cleanup happen, even if an error is raised. Files are the classic example. Lesson 36

context window

The maximum amount of text, measured in tokens, a language model can consider at once, including the whole conversation and its reply. Lesson 53

dataclass

A class decorated with @dataclass so Python writes its __init__, __repr__ and __eq__ from the field annotations. Lesson 33

decorator

A function that wraps another function to add behaviour, applied with @name above a definition. Lesson 35

dictionary

A collection of key-value pairs, looked up by key rather than by position. The container real programs are made of. Lesson 13

docstring

A string as the first line of a function, class or module, kept by Python as its documentation and shown by help(). Lesson 17

duck typing

Caring whether an object has the method you need, not what class it is. 'If it quacks like a duck.' Lesson 32

embedding

A list of numbers representing the meaning of a piece of text, so that similar meanings have similar vectors. The engine of semantic search. Lesson 58

environment variable

A value stored outside your code, in the operating system's environment, used to keep secrets like API keys out of your files. Lesson 54

exception

An error raised at runtime that stops normal flow, handled with try and except. Lesson 22

f-string

A string prefixed with f whose {braces} are replaced by the values inside them. The one obviously correct way to build text from values. Lesson 4

float

A number with a decimal point. Stored in binary, so tiny rounding errors are normal; use Decimal for money. Lesson 3

function

A named, reusable block of steps. The single most important idea for keeping programs understandable. Lesson 17

generator

A function using yield that produces values lazily, one at a time, using almost no memory. Lesson 34

GIL

The Global Interpreter Lock: in standard CPython, only one thread runs Python bytecode at a time, so threads speed up waiting but not CPU-bound work. Lesson 39

hallucination

When a language model produces fluent but fabricated output. A property of predicting plausible text, not a bug that can be fully removed. Lesson 53

immutable

Unable to be changed after creation. Strings, tuples and frozen dataclasses are immutable; lists and dicts are not. Lesson 4

indentation

The leading spaces that define blocks in Python. Unlike most languages, the layout is the syntax. Four spaces, never tabs. Lesson 7

index

The position of an item in a sequence, counting from zero. x[0] is the first item, x[-1] the last. Lesson 4

int

A whole number. Python integers grow to any size and never overflow. Lesson 3

interpreter

The program that runs your Python. Standard Python is CPython; Pyodide is CPython compiled to run in a browser. Base Camp 3

JSON

A text format for structured data, borrowed from JavaScript, used by nearly every web API. Maps onto Python dicts and lists. Lesson 23

lambda

A tiny anonymous function of one expression, e.g. lambda x: x*2. For throwaway use as an argument, never assigned to a name. Lesson 37

list

An ordered, changeable collection. The default container in Python. Lesson 11

list comprehension

See comprehension. Lesson 16

local scope

The region inside a function where its own variables live. Names assigned in a function are local unless declared otherwise. Lesson 19

method

A function that belongs to an object, called with a dot, e.g. text.upper(). Lesson 4

module

A .py file of code you can import. The standard library is hundreds of them. Lesson 20

mutable

Able to be changed in place after creation. Lists, dicts and sets are mutable; the source of the aliasing surprise. Lesson 11

None

Python's value for 'deliberately nothing'. Returned by functions with no return; compared with is None. Lesson 6

PEP 8

Python's official style guide. Following it means any Python programmer can read your code without friction. Lesson 30

pip

The tool that installs packages from PyPI into your environment. Lesson 26

prompt injection

An attack where text the model reads contains hidden instructions aimed at it. Unsolved, and the central risk of giving a model tools. Lesson 62

property

A method used like an attribute via @property, so a value can be computed on demand or validated on assignment. Lesson 31

PyPI

The Python Package Index: half a million public packages installable with pip. Lesson 26

RAG

Retrieval-augmented generation: find the relevant piece of your own data and paste it into the prompt so a model can answer from information it was never trained on. Lesson 58

recursion

A function that calls itself. Needs a base case that returns without recursing, or it runs forever. Lesson 58

regular expression

A tiny language for describing shapes of text, used via the re module for search and replace. Lesson 25

REPL

The interactive Python prompt (Read, Evaluate, Print, Loop). A calculator that speaks Python; the best tool for 'what does this do?'. Base Camp 6

return

Hands a value back from a function to its caller. Different from print, which shows a value and returns nothing. Lesson 17

scope

Where a name is visible. Python searches Local, Enclosing, Global, Built-in, in that order (LEGB). Lesson 19

secrets

The standard module for cryptographically safe random values: tokens, passwords, keys. Never use random for security. Lesson 52

set

An unordered collection with no duplicates and instant membership testing. Lesson 14

shadowing (variable)

Reusing a name with a fresh let-style assignment, or accidentally hiding a built-in like list by assigning to that name. Lesson 19

slice

A piece of a sequence, x[start:stop], including the start and excluding the stop. Lesson 4

SQL injection

An attack where user input rewrites a database query. Prevented by parameterised queries with ? placeholders, never f-strings. Lesson 45

streaming

Displaying a model's reply token by token as it is generated, so the wait feels alive rather than frozen. Lesson 56

string

Text, written in quotes. A sequence of characters, and immutable. Lesson 4

system prompt

Standing instructions that set an assistant's persona and rules, applied to every turn, separate from the conversation. Lesson 54

temperature

A dial controlling how randomly a model samples its next token. Low is focused and predictable; high is creative and surprising. Lesson 53

token

The chunk of text a language model reads and predicts, roughly three-quarters of a word. You pay per token. Lesson 53

tool use

Letting a model ask your code to run functions (function calling), so it can act: check data, do maths, control things. Powerful and risky. Lesson 57

traceback

The report Python prints when an exception is not caught, naming what broke, where, and why. Read it from the bottom. Lesson 10

truthiness

How any value is treated as True or False: empty and zero are falsy, everything else is truthy. Lesson 6

try / except

The way to handle exceptions: attempt the risky code, catch the specific errors you can deal with. Lesson 22

tuple

An ordered, unchangeable collection. Documents that a fixed group of values belongs together. Lesson 12

type hint

An annotation like name: str that documents the expected type. Not enforced at runtime, but checked by tools like mypy. Lesson 38

unpacking

Assigning several names from a sequence at once, e.g. a, b = point, or spreading with a star. Lesson 12

variable

A label attached to a value, created with =. In Python the value has a type; the label does not. Lesson 2

virtual environment

A private, isolated set of installed packages for one project, so projects with conflicting versions never collide. Lesson 26

walrus operator

:=, which assigns a value and returns it in one expression, e.g. if (n := len(x)) > 3:. Lesson 41

Nothing matches that. Try a shorter search.