Emerging technologies, such as chatbots, offer potential solutions by enabling immersive, scenario based language practice. Unfortunately,most language educators lack the technical expertise to develop and integrate chatbot-driven exercises into their classrooms. Approaches like low-code and Model-driven Engineering promise to make this process more accessible, allowing non-technical experts to create tailored solutions. However, as technology rapidly evolves, the challenge is to provide tools that are both adaptable and sustainable, enabling educators to design and deploy digital exercises independently.

This thesis explores a solution that bridges the gap between the pedagogical needs of educators and the technical complexity of modern tools. Using Design Science Research (DSR), an artifact was developed and validated that simplifies the creation of specialised language exercises for high-risk operational domains. The artifact allows trainers to digitize their training materials, and enrich it with chatbot functionality on a conceptual modeling level without requiring programming skills. Ultimately, it enables technology-agnostic language trainers to deliver effective, context-specific language training that promotes more precise and more natural communication in critical railway operations. The study evaluates the overall user experience and the resulting quality of the artifact through qualitative feedback and tests.