diff --git a/python-mixins/README.md b/python-mixins/README.md
new file mode 100644
index 0000000000..d8a1b67c93
--- /dev/null
+++ b/python-mixins/README.md
@@ -0,0 +1,9 @@
+# TODO
+
+This folder contains the source code that accompanies the Real Python tutorial, [TODO](TODO).
+
+## Usage
+
+```shell
+$ DATABASE=database.db python main.py
+```
diff --git a/python-mixins/TODO.md b/python-mixins/TODO.md
new file mode 100644
index 0000000000..de6c004bda
--- /dev/null
+++ b/python-mixins/TODO.md
@@ -0,0 +1,2 @@
+- Add nullable foreign key handling
+- Implement examples to test and demonstrate
diff --git a/python-mixins/main.py b/python-mixins/main.py
new file mode 100644
index 0000000000..0e726802f9
--- /dev/null
+++ b/python-mixins/main.py
@@ -0,0 +1,49 @@
+import logging
+
+logging.basicConfig(
+    level=logging.DEBUG,
+    format="\N{rocket} %(message)s"
+)
+
+from orm import ActiveRecord, TimestampMixin, SQLMixin
+
+
+class User(ActiveRecord):
+    first_name: str
+    last_name: str
+    team: bool = False
+
+
+class Tutorial(SQLMixin, ActiveRecord):
+    author: User
+    title: str
+    url: str
+
+
+class Comment(TimestampMixin, SQLMixin, ActiveRecord):
+    text: str
+    tutorial: Tutorial
+    user: User | None = None
+
+
+
+def populate():
+    user = User("Joanna", "Jablonski", team=True)
+    user.save()
+
+    tutorial = Tutorial(
+        author=user,
+        url="https://realpython.com/python-f-strings/",
+        title="Python's F-String for String Interpolation and Formatting",
+    )
+    tutorial.save()
+
+    comment = Comment(
+        text="A neutral comment from an anonymous reader",
+        tutorial=tutorial,
+    )
+    comment.save()
+
+
+if __name__ == "__main__":
+    populate()
diff --git a/python-mixins/orm/__init__.py b/python-mixins/orm/__init__.py
new file mode 100644
index 0000000000..d5a202ee0f
--- /dev/null
+++ b/python-mixins/orm/__init__.py
@@ -0,0 +1,8 @@
+from orm.core import ActiveRecord
+from orm.mixins import SQLMixin, TimestampMixin
+
+__all__ = [
+    "ActiveRecord",
+    "SQLMixin",
+    "TimestampMixin",
+]
diff --git a/python-mixins/orm/core.py b/python-mixins/orm/core.py
new file mode 100644
index 0000000000..fb1b818be5
--- /dev/null
+++ b/python-mixins/orm/core.py
@@ -0,0 +1,306 @@
+import atexit
+import inspect
+import os
+import re
+import sqlite3
+import textwrap
+import typing
+from dataclasses import MISSING, Field, dataclass, field, fields
+from functools import cached_property
+from typing import Any, ClassVar, Iterator, Type, TypeVar
+
+DATABASE_FILE = os.getenv("DATABASE", ":memory:")
+PRIMARY_KEY_COLUMN = "pk"
+SQL_COLUMN_TYPES = {
+    bool: "NUMERIC",
+    bytes: "BLOB",
+    float: "REAL",
+    int: "INTEGER",
+    str: "TEXT",
+}
+
+_T = TypeVar("_T", bound="ActiveRecord")
+
+
+class DataClassMeta(type):
+    def __new__(mcs, name, bases, attrs, **kwargs):
+        cls = super().__new__(mcs, name, bases, attrs)
+        return dataclass(**kwargs)(cls)
+
+
+class ActiveRecordMeta(DataClassMeta):
+    connection = sqlite3.connect(DATABASE_FILE)
+    connection.execute("PRAGMA foreign_keys = ON")
+    connection.row_factory = sqlite3.Row
+    connection.autocommit = True
+    atexit.register(connection.close)
+
+    def __new__(mcs, name, bases, attrs):
+        cls = super().__new__(mcs, name, bases, attrs)
+        if len(bases) > 0:
+            cls.__table__ = SQLTable(cls)
+            cls.__table__.create()
+        return cls
+
+
+class ActiveRecord(metaclass=ActiveRecordMeta):
+    __table__: ClassVar["SQLTable"]
+    pk: int | None = field(kw_only=True, default=None)
+
+    @classmethod
+    def find_all(cls: type[_T]) -> Iterator[_T]:
+        return recursive_fetch(cls, cls.__table__.select_all())
+
+    @classmethod
+    def find_by(cls: type[_T], **parameters) -> Iterator[_T]:
+        if not parameters:
+            raise ValueError("missing query conditions")
+        return recursive_fetch(cls, cls.__table__.select_where(**parameters))
+
+    @classmethod
+    def find(cls: type[_T], *, pk: int) -> _T:
+        try:
+            return next(cls.find_by(pk=pk))
+        except StopIteration as ex:
+            raise ValueError(f"{cls.__name__} with pk={pk} not found") from ex
+
+    def save(self) -> None:
+        if self.pk is None:
+            cursor = self.__table__.insert(self)
+            self.pk = cursor.lastrowid
+        else:
+            self.__table__.update(self)
+
+    def delete(self) -> None:
+        self.__table__.delete(self)
+        self.pk = None
+
+    def __setattr__(self, name: str, value: Any) -> None:
+        if name == PRIMARY_KEY_COLUMN:
+            if frame := inspect.currentframe():
+                if calling_frame := frame.f_back:
+                    local_self = calling_frame.f_locals.get("self", None)
+                    if local_self is not self:
+                        raise AttributeError(
+                            "primary key is managed by the database"
+                        )
+        super().__setattr__(name, value)
+
+
+class SQLTable:
+    def __init__(self, cls: type[_T]) -> None:
+        self.cls = cls
+        self.sql = SQLQueryGenerator(self)
+
+    @cached_property
+    def name(self) -> str:
+        return f"{snake_case(self.cls.__name__).rstrip("s")}s"
+
+    @cached_property
+    def columns(self) -> list["SQLColumn"]:
+        return [
+            SQLColumn(class_field)
+            for class_field in fields(self.cls)
+            if class_field.name != PRIMARY_KEY_COLUMN
+        ]
+
+    @cached_property
+    def foreign_keys(self) -> dict[str, type[_T]]:
+        return {
+            column.name: column.foreign_table.cls
+            for column in self.columns
+            if column.foreign_table
+        }
+
+    def create(self) -> sqlite3.Cursor:
+        return self.cls.connection.execute(self.sql.create.statement)
+
+    def insert(self, record: _T) -> sqlite3.Cursor:
+        if record.pk is not None:
+            raise ValueError("record has a primary key")
+        query = self.sql.insert(record)
+        return self.cls.connection.execute(query.statement, query.values)
+
+    def update(self, record: _T) -> sqlite3.Cursor:
+        if record.pk is None:
+            raise ValueError("record has no primary key")
+        query = self.sql.update(record)
+        return self.cls.connection.execute(query.statement, query.values)
+
+    def delete(self, record: _T) -> sqlite3.Cursor:
+        if record.pk is None:
+            raise ValueError("record hasn't been saved to database")
+        query = self.sql.delete(record)
+        return self.cls.connection.execute(query.statement, query.values)
+
+    def select_all(self) -> sqlite3.Cursor:
+        return self.cls.connection.execute(self.sql.select_all.statement)
+
+    def select_where(self, **parameters) -> sqlite3.Cursor:
+        query = self.sql.select_where(**parameters)
+        return self.cls.connection.execute(query.statement, query.values)
+
+
+class SQLColumn:
+    name: str
+    type: str
+    default: Any
+    foreign_table: SQLTable | None
+
+    def __init__(self, class_field: Field) -> None:
+        field_type = primary_type(class_field.type)
+        if issubclass(field_type, ActiveRecord):
+            self.name = f"{class_field.name}_{PRIMARY_KEY_COLUMN}"
+            self.type = SQL_COLUMN_TYPES.get(int, "TEXT")
+            self.foreign_table = SQLTable(field_type)
+        else:
+            self.name = class_field.name
+            self.type = SQL_COLUMN_TYPES.get(field_type, "TEXT")
+            self.foreign_table = None
+        if class_field.default_factory is not MISSING:
+            self.default = class_field.default_factory()
+        elif class_field.default is not MISSING:
+            self.default = class_field.default
+        else:
+            self.default = MISSING
+
+    @cached_property
+    def definition(self) -> str:
+        sql = f"{self.name} {self.type}"
+        if self.foreign_table:
+            sql += (
+                " REFERENCES "
+                f"{self.foreign_table.name}({PRIMARY_KEY_COLUMN})"
+            )
+        if self.default is MISSING:
+            sql += " NOT NULL"
+        elif self.default is not None:
+            sql += f" DEFAULT {self.default!r}"
+        return sql
+
+
+class SQLQuery:
+    def __init__(
+        self, statement: str, parameters: dict[str, Any] | None = None
+    ) -> None:
+        self.statement = statement
+        self.parameters = parameters or {}
+        self.values = rename(self.parameters)
+
+    def __str__(self) -> str:
+        result = self.statement
+        for key, value in self.values.items():
+            if value is None:
+                result = result.replace(f"=:{key}", " IS NULL")
+            else:
+                result = result.replace(f":{key}", repr(value))
+        return result + ";"
+
+
+class SQLQueryGenerator:
+    def __init__(self, table: "SQLTable") -> None:
+        self.table = table
+
+    @cached_property
+    def create(self) -> SQLQuery:
+        column_definitions = [
+            f"{PRIMARY_KEY_COLUMN} INTEGER PRIMARY KEY AUTOINCREMENT",
+            *(column.definition for column in self.table.columns),
+        ]
+        return SQLQuery(
+            textwrap.dedent(
+                f"""\
+                CREATE TABLE IF NOT EXISTS {self.table.name}(
+                    {",\n                    ".join(column_definitions)}
+                )"""
+            )
+        )
+
+    def insert(self, record: _T) -> SQLQuery:
+        column_names = ", ".join(column.name for column in self.table.columns)
+        placeholders = ", ".join(
+            f":{column.name}" for column in self.table.columns
+        )
+        return SQLQuery(
+            (
+                f"INSERT INTO {self.table.name}({column_names}) "
+                f"VALUES ({placeholders})"
+            ),
+            dict(vars(record)),
+        )
+
+    def update(self, record: _T) -> SQLQuery:
+        placeholders = ", ".join(
+            f"{column.name}=:{column.name}" for column in self.table.columns
+        )
+        return SQLQuery(
+            (
+                f"UPDATE {self.table.name} "
+                f"SET {placeholders} "
+                f"WHERE {PRIMARY_KEY_COLUMN}=:{PRIMARY_KEY_COLUMN}"
+            ),
+            dict(vars(record)),
+        )
+
+    def delete(self, record: _T) -> SQLQuery:
+        return SQLQuery(
+            (
+                f"DELETE FROM {self.table.name} "
+                f"WHERE {PRIMARY_KEY_COLUMN}=:{PRIMARY_KEY_COLUMN}"
+            ),
+            dict(vars(record)),
+        )
+
+    @cached_property
+    def select_all(self) -> SQLQuery:
+        return SQLQuery(f"SELECT * FROM {self.table.name}")
+
+    def select_where(self, **parameters) -> SQLQuery:
+        values = rename(parameters)
+        conditions = " AND ".join(f"{param}=:{param}" for param in values)
+        return SQLQuery(
+            f"SELECT * FROM {self.table.name} WHERE {conditions}",
+            values,
+        )
+
+
+def recursive_fetch(cls: type[_T], cursor: sqlite3.Cursor) -> Iterator[_T]:
+    for row in cursor.fetchall():
+        attrs = {}
+        for column_name in row.keys():
+            if fk := cls.__table__.foreign_keys.get(column_name):
+                name = column_name.removesuffix(f"_{PRIMARY_KEY_COLUMN}")
+                attrs[name] = fk.find(pk=row[column_name])
+            else:
+                attrs[column_name] = row[column_name]
+        yield cls(**attrs)
+
+
+def rename(parameters: dict[str, Any]) -> dict[str, Any]:
+    values = {}
+    for key, value in parameters.items():
+        if isinstance(value, ActiveRecord):
+            values[f"{key}_{PRIMARY_KEY_COLUMN}"] = value.pk
+        else:
+            values[key] = value
+    return values
+
+
+def primary_type(type_hint: Type) -> type:
+    if args := typing.get_args(type_hint):
+        match [arg for arg in args if arg is not type(None)]:
+            case [x] if x is not None:
+                return x
+            case []:
+                raise TypeError("no primary types found")
+            case _:
+                raise TypeError("cannot have multiple primary types")
+
+    if type_hint is None:
+        raise TypeError("type cannot be None")
+
+    return type_hint
+
+
+def snake_case(text: str) -> str:
+    return re.sub(r"([a-z])([A-Z])", r"\1_\2", text).lower()
diff --git a/python-mixins/orm/mixins.py b/python-mixins/orm/mixins.py
new file mode 100644
index 0000000000..9f7253c5e5
--- /dev/null
+++ b/python-mixins/orm/mixins.py
@@ -0,0 +1,57 @@
+import logging
+import time
+from dataclasses import field
+
+from orm.core import ActiveRecordMeta
+
+CREATED_AT_COLUMN = "created_at"
+UPDATED_AT_COLUMN = "updated_at"
+
+logger = logging.getLogger(__name__)
+
+
+class SQLMixin:
+    @classmethod
+    def find_all(cls):
+        logger.debug(cls.__table__.sql.select_all)
+        return super().find_all()
+
+    @classmethod
+    def find_by(cls, **parameters):
+        logger.debug(cls.__table__.sql.select_where(**parameters))
+        return super().find_by(**parameters)
+
+    def save(self):
+        if self.pk is None:
+            logger.debug(self.__table__.sql.insert(self))
+        else:
+            logger.debug(self.__table__.sql.update(self))
+        super().save()
+
+    def delete(self):
+        logger.debug(self.__table__.sql.delete(self))
+        return super().delete()
+
+
+class TimestampMixinMeta(ActiveRecordMeta):
+    def __call__(cls, *args, **kwargs):
+        created_at = kwargs.pop(CREATED_AT_COLUMN, None)
+        updated_at = kwargs.pop(UPDATED_AT_COLUMN, None)
+        instance = super().__call__(*args, **kwargs)
+        instance.created_at = created_at
+        instance.updated_at = updated_at
+        return instance
+
+
+class TimestampMixin(metaclass=TimestampMixinMeta):
+    created_at: int = field(init=False, repr=False)
+    updated_at: int = field(init=False, repr=False)
+
+    def save(self) -> None:
+        current_time = int(time.time())
+        if self.pk is None:
+            self.created_at = current_time
+            self.updated_at = current_time
+        else:
+            self.updated_at = current_time
+        super().save()