select(sort(
    filter(table(users), \ u: u.name == $1),
    \ u: u.id))

or:

  table(users) filter(name == $1) sort(id, ASC)
  table(users) filter(id == $1) update(salary = 10000)

select(join(table(user), table(department), \ u, d: d.id == u.department_id))

insert(table(user), {
  id: 1,
  name: "Foo"
  })

Low-level intermediate representation?

SELECT_FROM users: id, name
FILTER
  LOAD_COLUMN users.name
  LOAD_PARAMETER $1
  EQ
  RET
SORT users.id ASC