“How Does This Thing Work?”: Mapping Citizen Developer User Concepts to Underlying Software Programming Technical Capabilities in Generative AI for Entrepreneurship and Innovation

No-code development platforms integrate Large Language Models (LLMs) to provide opportunity for non-technical users, like entrepreneurs, to innovate in software development. However, they still necessitate a level of coding proficiency to use effectively. A case study of building a travel mobile app with low and high tech prompts in an LLM-based no-code development platform explores how to support non-technical users by mapping technical capabilities to product needs described in plain English. The low tech prompts yield variable results in successful feature implementation and can mimic functionality instead of using the needed technical capabilities. The findings from this evaluation inform prompting guidelines, supporting non-technical users in executing their business and product needs. Mapping user concepts to underlying technical capabilities can be used as a framework to support usability of generative artificial intelligence (GenAI) for programming applications.