The exponentially improving nanofabrication capabilities are one of the driving forces behind the technological advancements of the last 100 years, affecting virtually every area of human endeavour and enabling the advent of the Information Age. As the fabrication technologies are reaching the atomistic limits, the development of methods with not only high resolution, but also increasing functionality and flexibility are required to continue the technological progress. In particular, controlling the fabrication in three dimensions (3D) has proven to be a challenging task for the conventional fabrication methods. However, the capability of freely realizing 3D devices is paramount for the development of many areas of nanotechnology. This is particularly important in the field of nanomagnetism where the material properties are strongly dependent on its geometry and topology. Controlling these properties and expanding spintronic technologies to 3D has a great potential for applications in the future low-power nanoelectronic architectures.

In this thesis, Focused Electron Beam Induced Deposition (FEBID) is investigated as a 3D nanofabrication tool with unique capabilities for prototyping high-resolution complex 3D structures. The competing effects present during the deposition have been studied and the most significant deposition parameters for the common precursors have been determined. Based on these parameters, a computationally solvable theoretical model of the 3D FEBID deposition has been developed. Furthermore, a calibration procedure has been designed for in-situ measurement of these parameters. The model and the calibration procedure have been implemented in a 3D printing algorithm that is capable of realizing general 3D geometries directly from standard 3D printing files. The algorithm has been extensively tested with different precursors and structure designs, demonstrating the effectiveness of both the model and the calibration procedure.

The application of the developed 3D fabrication capabilities in nanomagnetism has subsequently been investigated. In particular, the effect of 3D geometry on the magnetic domain walls and their dynamics have been studied. The purely geometry-driven domain wall motion in the fabricated 3D domain wall conduits has been experimentally demonstrated, offering a route towards fast, robust and energy efficient route for transferring magnetic information between functional planes. Furthermore, micromagnetic studies of highly curved 3D domain wall conduits have been performed, showing that the domain wall energy and stability are strongly affected by the geometry, and in particular curvature, and can result in topological domain wall transformations. These studies present a route towards controlling the properties of domain walls and their dynamics in the future spintronic technologies.