Water is one of the biggest challenges that humans face. Either there is too little water, too much water or the quality of the water is poor. Management of water is the key solution to many of the problems, which the UN SDG are addressing. Existing methods used within this field are often based on passive strategies, where engineering designs and solutions rely on the evaluation of different scenarios based on a worst-case approach. During the VILLUM Synergy project CLAIRE we have demonstrated that using reinforcement learning it is possible to significantly improve measurement, prediction, control and regulation of water using active strategies.