The advances in multi-physics and multi-scale scientific simulations are the incentive for research in new ways of handling the development of complex simulation codes. The paradigm of component programming and Grid computing set a new level of requirements for simulation codes in terms of the interaction between highly heterogeneous components and the adoption of new codes. This thesis stresses the need for the development of software integratio~ techniques for scientific simulation codes. Ensuring component interoperability allows not only the building of more powerful programs but, as it, is strongly stressed in this thesis, helps to lower the cost of the verification and validation of simulation programs. This thesis introduces the notion of a hybrid simulation system as a system consisting of generic system programming language libraries, scripting language in. terpreter, interface modules between scripting language and system language components and interface generation' tools. It is argued that hybrid are to be the most appropriate environment for the development of academic simulation codes. The main contribution of this thesis is the idea of Grid and Geometry Exchange Services (GAGES) as an example of a hybrid system in the domain of pre- and postprocessing of scientific simulations. The case studies undertaken to support claims about GAGES and hybrid systems yielded several practical results, for instance an anisotropic mesh generator, a surface mesh generator, a grid plotting library and new tools for multi-language programming. A posteriori analysis of the development efforts resulted in another original idea of the usage of a SWIG compiler interface specification as a universal scientific interface description language.