Open data is widely recognized for its potential positive impact on society and economy. However, many open data sets remain underutilized because users, such as civil servants and citizens, lack the necessary technical and analytical skills. Additionally, existing open data portals often fall short of providing user-friendly access to data. Although conversational agents equipped with Large Language Models have emerged as a promising solution to address these challenges, it is unclear how to design Large Language Model based open data assistants that allow users to formulate their information needs in natural language and ultimately use open data effectively. To address this gap, we undertake a Design Science Research project guided by the theory of effective use. In this first cycle of the project, we present meta-requirements and propose initial design principles on how to design a Large Language Model based open data assistant for effective use. Subsequently, we instantiate our principles in a prototype and evaluate it in a focus group with experts from a medium-sized German city. Our results contribute design knowledge in the form of design principles for open data assistants and inform future design cycles of our Design Science Research project.