The availability of numerous algorithms and techniques developed for complex data analysis makes it difficult for domain users to navigate the life cycle workflow for their analysis. One way to overcome this is by providing interactive tools that are intuitive to use, graphical, and can be used for all stages of the life cycle by non-experts – from modeling to complex analysis to drill down & visualization. If these tools are also extensible and modular, they allow additions of new models and algorithms as they become available for the benefit of the larger community. Graphs have been extensively used to study the complex systems of interacting entities from diverse disciplines. However, when studying complex data with multiple types of entities and relationships, simple or even attributed graphs are not always ideal for modeling them. Currently, to derive knowledge from complex data with multiple entity types and relationships, multilayer networks (or MLNs) are being increasingly used. MLN-geeWhiz, is conceived to be such an end-to-end interactive tool with its initial version available for public use. Its purpose is to fill a void and allow non-experts from diverse disciplines to analyze their complex data using state-of-the-art analysis techniques and visualize the results. This dashboard has been developed to use complex analysis techniques with ease. It is modular, so new algorithms and approaches can be added transparently. The user can upload, generate needed MLNs (or graphs), analyze them and visualize results in many ways. In this paper, we discuss the challenges of a modular architecture needed for supporting complex analysis underneath and visualization to understand the results.