Electrocatalysis presents novel pathways for advanced water treatment. For example, electrocatalytic reduction is an emerging technology for treating oxyanions of concern in water. However, the identification of highly performant, cost-effective catalysts remains a major barrier to deployment at scale. This article discusses how computational modeling and machine learning (ML) can accelerate the search for new catalyst materials for electrocatalytic water treatment processes, such as the electroreduction of oxyanions and degradation of persistent organics. It describes how traditional computational chemistry workflows, now deployed in their basic form for at least two decades, can be expanded in breadth and depth through newly developed machine-learned force fields that have been trained on millions of data points. It also discusses ways in which the theory and ML pipeline can be effectively integrated with experimental synthesis and characterization platforms to rapidly identify and validate new catalyst chemistries for water purification challenges.