This paper presents the design and implementation of BL.MixedR, a maintenance assistance system that leverages the capabilities of augmented reality (AR) to assist technicians during their maintenance processes. BL.MixedR re-imagines work orders to take full advantage of the contextualization of digital data and maintenance information using the spatial and hands-free capabilities offered by augmented reality headsets. The prototype provides technicians with relevant information at the right time and place. This allows them to focus solely on the maintenance work, reducing unnecessary cognitive load and improving their performance. Moreover, the architecture of the BL.MixedR system allows for a simplified integration with existing maintenance systems, which we believe is one of the main barriers to the industrial adoption of AR maintenance systems. We present in this paper the result of several iterations of user-centered design and multiple evaluations by maintenance experts in real field conditions. These in-situ observations lead us to identify several empirical design principles to maximize the adoption of augmented reality systems during maintenance operations. CCS CONCEPTS •Human-centered computing~Human computer interaction (HCI)~Interaction paradigms~Mixed / augmented reality