AI-Augmented CAD Onboarding: A Personalized Approach to Reducing Learning Friction

Abstract

Autodesk Fusion is a leading cloud‑based CAD platform, yet new users often face steep learning curves due to scattered resources, inconsistent guidance, and a lack of personalization. This thesis addresses these challenges and proposes an adaptive AI assistant, embedded within Fusion, as a potential solution to streamline onboarding, reduce search time, surface hidden tools, and deliver guidance tailored to the user’s learning style. By centralizing learning support within the design environment, the proposed system aims to reduce cognitive load and keep users focused on productive work rather than on searching for help. Based on surveys, interviews, and controlled user testing comparing tasks with and without simulated AI support, the study suggests that personalized, context‑aware assistance can improve task flow, reduce frustration, and provide particular benefits for beginners. Findings indicate that such a solution not only accelerates skill acquisition but also supports long‑term engagement by making the early stages of learning more intuitive and less discouraging. Finally, this thesis outlines practical next steps Autodesk can take to develop, integrate, and validate such a system to realize its full potential in accelerating adoption, improving retention, and enhancing the overall user experience.