How was the work with databases carried out in the projects — through ORM or native tools?

The projects used a mixed approach: for standard operations and rapid prototyping, ORM was employed, which accelerated development and simplified code maintenance. However, for complex queries with high load and performance optimization, native SQL queries and stored procedures were used. This approach allowed balancing between development convenience and database efficiency.

Example using ORM in Python (SQLAlchemy):

session.query(User).filter(User.age > 30).all()

And a native SQL query for complex selection:

SELECT u.id, u.name, COUNT(o.id) as orders_count
FROM users u
LEFT JOIN orders o ON u.id = o.user_id
GROUP BY u.id, u.name
HAVING COUNT(o.id) > 5;