The absence of a suitable standard device platform for terahertz waves is currently a major roadblock that is inhibiting the widespread adoption and exploitation of terahertz technology. As a consequence, terahertz-range devices and systems are generally an ad hoc combination of several different heterogeneous technologies and fields of study, which serves perfectly well for a once-off experimental demonstration or proof-of-concept, but is not readily adapted to real-world use case scenarios. In contrast, establishing a common platform would allow us to consolidate our design efforts, define a well-defined scope of specialization for 'terahertz engineering,' and to finally move beyond the disconnected efforts that have characterized the past decades. This tutorial will present arguments that nominate substrateless all-silicon microstructures as the most promising candidate due to the low loss of high-resistivity float-zone intrinsic silicon, the compactness of high-contrast dielectric waveguides, the designability of lattice structures, such as effective medium and photonic crystal, physical rigidity, ease and low cost of manufacture using deep-reactive ion etching, and the versatility of the many diverse functional devices and systems that may be integrated. We will present an overview of the historical development of the various constituents of this technology, compare and contrast different approaches in detail, and briefly describe relevant aspects of electromagnetic theory, which we hope will be of assistance.