A regular tetrahedron can be inscribed in a cube in a way that the tetrahedron edges are diagonals of the cube faces.
Is it possible to similarly inscribe any irregular tetrahedron in a cuboid or prism?
Image source: http://mathworld.wolfram.com/RegularTetrahedron.html
Affine transformations of the cube allow inscribing absolutely any tetrahedron in a parallelepiped, the unit cell of the triclinic crystal system. Now, for the three opposite pairs of edges of a tetrahedron, consider the vectors formed as cross products of each pair. Extending the previous line of reasoning, we see that