High-energy particle physics is going through a crucial moment of its history, one in which it can finally aspire to give a precise answer to some of the fundamental questions it has been conceived for. On the one side, the theoretical picture describing the elementary strong and electroweak interactions below the TeV scale, the Standard Model, has been well consolidated over the decades by the observation and the precise characterization of its constituents. On the other hand, the enormous technological potentialities nowadays available, and the skills accumulated in decades of collider experiments with increasingly high complexity, render for the first time plausible the possibility of addressing complicated and conceptually deep questions like the ones at hand. The best incarnation of this high level of sophistication is the CERN Large Hadron Collider (LHC), the most powerful experimental apparatus ever built, which is designed to shed light on the true nature of fundamental interactions at energies never attained before, and which has already started to open a new era in physics with the recent discovery of the longed-for Higgs boson, a true milestone for the human knowledge as well as one of the most important discoveries in the modern epoch. The knowledge that has been and is going to be reached in these crucial years would of course not be conceivable without a deep interplay between the theoretical and the experimental efforts. In particular, on the theoretical side, not only there are wide groups of researchers devoted to building possible extensions to the Standard Model, which draws the guidelines of current and future experiments, but also there is a vast community whose research is rather aimed at the precise predictions of all the physical observables that could be measured at colliders, and at the systematic improvement of the approximations that currently constrain such predictions. On top of representing the state-of-the-art of the human understanding of the properties that regulate elementary-particle interactions and of the formalisms that describe them, the developments of this line of research have an immediate and significant impact on experiments. Firstly, these detailed calculations are the very theoretical predictions against which experimental data are compared, so they are crucial in establishing the validity or not of the theories according to which they are performed. Secondly, the signals one wants to extract from data at modern colliders are so tiny and difficult to single out that the experimental searches themselves need be supplemented by a detailed work of theoretical modelling and simulation. In this respect, high-precision computations play an essential role in all analysis strategies devised by experimental collaborations, and in many aspects of the detector calibration. It is clear that, for theoretical computations to be useful in experimental analyses and simulations, the predictions they yield should be reliable for all possible configurations of the particles to be detected. Thus the key feature for the present theoretical collider physics is not particularly the computation of observables with high precision only in a limited region of the phase space, but the capability of combining (‘matching’) in a consistent way different approaches, each of which is reliable in a particular kinematic regime. With this perspective, matching techniques represent one of the most promising and successful theoretical frameworks currently available, and are considered as eminently valuable tools both on the theoretical and on the experimental sides. Matched computations are based on a perturbation-theory approach for the description of configurations in which the scattering products are well separated and/or highly energetic: in particular the precision currently attained for all but a few of the relevant processes within the Standard Model is the next-to-leading order (NLO) in powers of the strong quantum-chromodynamics (QCD) coupling constant αS; for the description of configurations in which the particles outgoing the collisions are close to each other and/or have low energy, it can be shown that the perturbation-theory expansion breaks down, and then a complementary method, like the parton shower Monte Carlo (PSMC), has instead to be employed. The task of matching is precisely that of giving a prediction that interpolates between the two approaches in a smooth and theoretically-consistent way. This thesis is focused on MC@NLO, a high-energy physics formalism capable of matching computations performed at the NLO in QCD to PSMC generators, in such a way as to retain the virtues of both approaches while discarding their mutual deficiencies. In particular, the thesis reports on the work successfully achieved in extending MC@NLO from its original numerical implementation, tailored on the HERWIG PSMC, to the other main PSMC programs currently employed by experimental collaborations, PYTHIA and Herwig++, confirming the advocated universality of the method. Differences in the various realizations are explained in detail both at the formal level and through the simulation of various Standard-Model reactions. Moreover we describe how the MC@NLO framework has been developed so as to render its implementation automatic with respect to the physics process one is about to simulate: beyond yielding an enormous increase in its potential for present and future collider phenomenology, and upgrading the standard of precision for high-energy computations to the NLO+PSMC level, this development allows for the first time the application of the MC@NLO formalism to a huge number of relevant and highly complicated reactions, through an implementation which is also easily usable by people well-outside the community of experts in QCD calculations. As example of this new version, called aMC@NLO, recent results are presented for complex scattering processes, involving four or five final-state particles. Finally, possible extensions of the framework to theories beyond the Standard Model, like the supersymmetric version of QCD, are briefly introduced.