Language use, second-language development, and technology mediated human activity are complex processes situated in, and in some cases demonstrably interwoven with, specific material and social contexts. This study highlights the context embedded and context producing interactional practices of learning in the wild as participants in small groups notice visible aspects of their immediate environment. The groups are involved in mobile augmented reality (AR) game play and are walking across an urban university campus and adjacent environments. Video-recorded interactions from 15 groups of three participants from four languages (English, German, Hungarian, and Japanese) were observed and transcribed. Sequential, multimodal analysis revealed numerous instances of noticing environmental resources and we show how participants use coordinated gaze, gesture, and language to make relevant particular perceived objects from the built environment for accomplishing the groups’ goal-directed activity as well as for co-constructing socio-environmental infrastructures for learning.