Is the SQLAlchemy Core — SQL as Python lesson free?

Yes. The first 5 lessons on PyRun are free forever — no signup, no card. SQLAlchemy Core — SQL as Python runs real Python in your browser via Pyodide.

How long does the SQLAlchemy Core — SQL as Python lesson take?

About 15 minutes end-to-end. It's a hands-on lesson — you read a paragraph, edit the code, and run it in the same tab.

Do I need to install Python to try the SQLAlchemy Core — SQL as Python lesson?

No install required. PyRun runs real CPython in your browser via Pyodide (Python compiled to WebAssembly). Works on Chrome, Firefox, Safari, and Edge — desktop or mobile.

What level is the SQLAlchemy Core — SQL as Python lesson?

Advanced level. Assumes comfort with classes, functions, and standard library. Not for first-time Python learners.

← All lessons

Advanced·15 min·data · sql · sqlalchemy · database · orm

SQLAlchemy Core — SQL as Python

Raw SQL strings are hard to compose and easy to get wrong. SQLAlchemy Core gives you an expression language — write queries as Python objects, get real SQL out.

Why bother

Type-checked in your editor — users.c.email == "..." is autocompleted.
Composable — filters, joins, aggregates are Python expressions you can build up conditionally.
Portable — same code runs against SQLite, Postgres, MySQL, MSSQL.
No injection — parameters are bound automatically.

Core primitives

create_engine("sqlite:///:memory:") — connection pool + dialect.
Table(name, MetaData(), Column(...)) — describe your schema.
select(t).where(...).order_by(...) — build a SELECT.
insert(t) / update(t) / delete(t) — the other DML verbs.
with engine.begin() as conn: — auto-commits on success, rolls back on exception.

When to use Core vs the ORM

Core: ETL, analytics, bulk operations, complex reporting queries.
ORM (next lesson): object-heavy business logic with rich relationships.
Both are the same library — you can mix them freely.

Try it

Add a query that counts users per plan.
Update Bob's plan to "pro" and re-run the select.

Practice

3 graded exercises

Write the code, click Check. We’ll tell you exactly what to fix.

Progress0 / 3

Exercise 1
Create an in-memory SQLite engine and define a users table with two columns: id (Integer, primary key) and email (String, unique). Then create the table in the database by calling meta.create_all(engine).
Exercise 2
The engine, metadata, and a users table are already set up. Inside a with engine.begin() as conn: block, insert two rows with emails 'alice@x.com' and 'bob@x.com'. Then run select(users) and print each row so both emails appear in the output.
Exercise 3
The users table already has 'alice@x.com' and 'bob@x.com' inserted. Write a select(users).where(...) query that finds the row where email equals 'alice@x.com', then print only that row's id.

← PreviousSQL joins that actually make sense Next →SQLAlchemy ORM — objects that persist

SQLAlchemy Core — SQL as Python

Raw SQL strings are hard to compose and easy to get wrong. SQLAlchemy Core gives you an expression language — write queries as Python objects, get real SQL out.

Why bother

Type-checked in your editor — users.c.email == "..." is autocompleted.
Composable — filters, joins, aggregates are Python expressions you can build up conditionally.
Portable — same code runs against SQLite, Postgres, MySQL, MSSQL.
No injection — parameters are bound automatically.

Core primitives

create_engine("sqlite:///:memory:") — connection pool + dialect.
Table(name, MetaData(), Column(...)) — describe your schema.
select(t).where(...).order_by(...) — build a SELECT.
insert(t) / update(t) / delete(t) — the other DML verbs.
with engine.begin() as conn: — auto-commits on success, rolls back on exception.

When to use Core vs the ORM

Core: ETL, analytics, bulk operations, complex reporting queries.
ORM (next lesson): object-heavy business logic with rich relationships.
Both are the same library — you can mix them freely.

Try it

Add a query that counts users per plan.
Update Bob's plan to "pro" and re-run the select.

Practice

3 graded exercises

Write the code, click Check. We’ll tell you exactly what to fix.

Progress0 / 3

Exercise 1
Create an in-memory SQLite engine and define a users table with two columns: id (Integer, primary key) and email (String, unique). Then create the table in the database by calling meta.create_all(engine).
Exercise 2
The engine, metadata, and a users table are already set up. Inside a with engine.begin() as conn: block, insert two rows with emails 'alice@x.com' and 'bob@x.com'. Then run select(users) and print each row so both emails appear in the output.
Exercise 3
The users table already has 'alice@x.com' and 'bob@x.com' inserted. Write a select(users).where(...) query that finds the row where email equals 'alice@x.com', then print only that row's id.