Higher education in traditional, highly centralized institutions is struggling to adequately prepare professionals in accordance to the current job environment, where demonstrating abilities to quickly adapt to changes and learn new skills is progressively gaining weight against solely holding a college degree, for example. Elevated tuition costs is another factor pushing people into short, internet-based specialized courses or directly into the job market, while an increasing number of people holding full degrees are underemployed in respect to their knowledge level. We propose VUZ, a domain-general framework capable of tackling these issues by designing personalized curricula from a high-level perspective (general courses, subjects and topics) down to specific how to's and step-by-steps, by means of crowdsourcing: delegating the curriculum development to a community of experts. The framework is scalable in the sense that individuals can choose the level of refinement of their curriculum depending on their prior knowledge; is decentralized in the sense that all decisions are taken by the users; and is collaborative because of the possibility of sharing knowledge with other users. We detailed the crowdsourcing algorithm used for curricula development and simulated a VUZ platform using a multiagent system, which yielded a series of cost and retribution results associated to potential users of a VUZ system. © 2018 IEEE.