Machine Learning (ML) has become one of the most promising technological advances for enterprises to improve manual, highly resource- and time-consuming processes. Developing and deploying these ML based systems in an organizational setting, however, is linked to a range of processual and technical requirements and implications that researchers and enterprises have only started to comprehend. Based on an Action Design Research approach, this study develops a ML based solution for data quality (DQ) controls, an essential instrument in Data Quality Management. We synthesize our findings through a set of design principles for ML based DQ controls that describe key components in the three phases from proof-of-concept to deployment and business process integration. Our findings lay groundwork for future research in the field of ML based systems for DQ and contribute to the broader IS discourse on how to embed learning-based systems in real-world organizational contexts.