Providing robust and scalable computational infrastructures to support data science processes and algorithms has become a fundamental requirement for generating valuable knowledge from large volumes of data, regardless of the user type or application domain. In this context, the convergence of cloud computing, container technology, and Model-Driven Engineering (MDE) emerges as an innovative and still underexplored paradigm for enabling such computational provisioning in an efficient, cost-effective, and accessible manner, especially for data owners with limited technical knowledge and resources in this area. Therefore, this paper proposes a Domain-Specific Language (DSL) and a model-based transformation engine aimed at supporting data experts in the visual design and automatic implement of multi-container infrastructures for data-intensive applications. The empirical evaluation of the solution, conducted through a quasi-experiment based on the Methods Evaluation Model (MEM), demonstrated participants’ willingness to adopt it in future projects, as they perceived it to be both easy to use and useful.