AI-Driven Reshaping of Product-Development Workflows: From Idea to Product

Sammanfattning

This thesis examines how AI developer-support tools reshape product-developmentworkflows in startups, scaleups and established firms. It focuses on vibe coding, agentic AItools and their effects on the build-measure-learn cycle and the early idea-to-product process.Based on 15 semi-structured interviews with respondents in Sweden, the findings show thatAI most clearly accelerates the build phase by lowering the time, cost and technical thresholdof creating prototypes and MVPs. However, faster building only creates value if teams stillvalidate, interpret and act on feedback. The findings also show that organisational contextshapes whether AI-enabled speed becomes product-level acceleration. Startups capture speedthrough autonomy, scaleups through deeper integration and established firms are moreconstrained by security, governance and approval processes. The thesis contributes byshowing how AI reshapes knowledge barriers, experimentation speed and organisationaladvantage.