The original vision of HyPer was to build an e cient combined OLTP and OLAP system by exploiting modern hardware, in particular large main-memory capacities. This original design led to a number of further developments, inspired by both technology changes and a broadening of the original goal. Key technologies were a powerful compilation framework and a highly parallelized query processing engine. This talk highlights the evolution and major development steps of HyPer from a specialized research system to an industrial-strength database system.