Abstract: Embodiments presented herein provide techniques for generating a collage of multimedia (e.g., visual and auditory media) based on data associated with a user. The collage may be used as an audio/visual messaging sharing vehicle. A platform engine receives user profile information and a selection of a performance mode from a client device. The platform engine generates a set of calls for user action prompts (e.g., questions or other requests for input) based on the selected performance mode and user information. The platform engine sends the set of prompts to the client device, and upon receiving responses to the prompts from the client device, the platform engine correlates the responses and the user profile information with media items in the media library. Based on the correlation, the platform engine generates the multimedia collage and sends the collage to the client device.