Higher Education Institutions (HEIs) have been criticised by employers, government and graduates themselves, for not adequately developing required work skills. An example of practice that does develop student skills is a short industrial placement (SIP) where students are expected to solve a real problem in a company, in two weeks, working with one other student. This practice occurs in a one year Masters programme at Cambridge University Engineering Department. This work studies the SIP practice to understand why it is effective and determine lessons that could contribute to solving the wider skills problem. A five year research timeframe, coupled with an annually run programme, enabled a multi-stage study using an Engaged Scholarship methodology. The first-stage was an exploratory study that investigated the initial development of SIP skills, using simulated experiences, in a taught HE based module. Skills development was found to be a complex multi-component process. A theoretical skills development framework was constructed from literature and compared with practice. It was determined that five simulated SIP experiences provided the student with sufficient skills to undertake a SIP in practice and, the most significant problem was that SIP skills were not well defined. The second-stage focussed on defining skills. Skills were found to be context specific and defining skills required both the associated task and its context to be known. With tasks found to be both essential to defining skills and effective in describing what graduates do in practice, a SIP task framework was constructed which was tested on 80 different SIPs in one academic year. The resulting framework comprised twelve problem-solving process-stages, that in total contained 64 different tasks, and five generic task domains. These generic domains were investigated in the third-stage of this research. These were found to be more extensive and complex than anticipated resulting in a reconfiguration of the SIP framework, the generation of SIP specific domain descriptions and partial completion of task frameworks to describe each domain. This research has generated a plausible skills development theory for HEIs, and task frameworks to describe a SIP. Further work has been identified to refine the task frameworks and to continue work on the proposed skills development theory.