Is Venus upside down? Does its north point in the same direction as Earth's south? (roughly) Or does it just spin clockwise and its north points towards Earth's north? (also roughly) It seems that the right-hand rule would point Venus north in the same direction Earth's south (roughly) but some have said Venus is not upside down, which I believe would mean Venus' north points (rougly) in the same direction as Earth's north.
This is a matter of defintion and convention, not science.
By convention "North pole" is defined to be the pole that lies to north side of the solar system's "invariable plane" (that is the plane in which the planets orbit). And so, by definition, its North Pole points the same way as Earth's North pole (roughly).
However axial tilt is defined relative to the positive pole, as defined by the right-hand-rule. For Earth and five other planets, the positive pole is the North pole. For Venus and Uranus the positive pole is the South pole, so these planets have inclinations that are greater than 90 degrees.
But remember that these are just arbitrary choices of nomenclature.
See this explanation on astronomy.com
This is a matter of convention and definition, not science (as @jamesk states).
It depends what "north" means. Presumably, you are not asking about magnetic north.
Generally, the celestial north pole is defined relative to a spinning body as the celestial pole in the positive direction along it's axis of rotation, applying the right-hand rule as normal. Celestial poles are the points on the surface of the body intersected by its axis of rotation. This matches your description. Earth and Venus's celestial poles do not align relative to their respective surfaces, and change relative to each other as their relative orbital positions change. Celestial north is used for describing satellites in Earth orbit and is nearly the same as magnetic north, so is usually what "north" means without additional context.
Orbital poles are the surface points of an orbiting object which are intersected by a line through that object's center of gravity and perpendicular to its orbit. The north one is the one in the positive direction when applying the right-hand rule to the object's motion relative to the point it is orbiting (usually a larger object, but it works for two similarly-sized objects as well). This is essentially @jamesk's answer. Because Earth and Venus orbit in essentially the same plane, their orbital poles are essentially parallel. Orbital poles are non-intuitive as applied to a planet, because they sweep through a latitude over the course of about a day. They are much more convenient when discussing relative planetary positions though, and so can also be a natural answer to your question.
I'm going to disagree with the other two answers in a particular way.
There is a way to take the "roughness" out of this question, and make this question more precise and scientific, not just definition and convention.
Namely, assuming by approximation that one has an inertial coordinate system for the solar system, the axis of each of Venus and Earth can be oriented in such a way that the spin around the axis and the orientation of the axis obey the right hand rule. One can then measure the angle between those two oriented axes. The numerical value of this angle will not depend on which inertial coordinate system you picked.
One could actually do this measurement: put a gyroscope in your spaceship with its axis oriented to have $0^\circ$ angle with the Earth's axis; then fly to Venus and observe the angle between your gyroscope axis and Venus' axis.
To say that Venus is "upside down" with respect to earth would then mean that this angle is $180^\circ$, to within some acceptable error.