The development of digital technology has changed the communication patterns of Indonesian society, especially among the younger generation of social media users. The phenomenon of code-mixing between Indonesian and English has become a form of linguistic hybridity that reflects the social dynamics and digital identity of the younger generation. This study aims to describe the patterns, pragmatic functions, and social implications of code-mixing practices on social media. The approach used is a qualitative descriptive approach with a critical sociolinguistic paradigm. Data were obtained through digital observation (netnography) of public posts on Instagram, Twitter (X), and TikTok during January–June 2025, with a total of 120 data units that met the purposive sampling criteria. The analysis was conducted using the model of Miles, Huberman, and Saldana (2014). The results show three dominant forms of code-mixing: at the word level (40%), phrases (32.5%), and sentences (27.5%). The most prominent pragmatic functions are the expression of identity and style (34%) and the affirmation of meaning (28%). Socially, code-mixing functions as a performative strategy to build a digital self-image and negotiate the global-local identity of the younger generation. This phenomenon marks a shift in communication norms toward more expressive, inclusive, and global forms. Thus, code-mixing is not seen as a threat to the purity of the Indonesian language but rather as a form of creative adaptation to the dynamics of digital communication. This research contributes to the development of digital sociolinguistic studies and offers a basis for Indonesian language preservation strategies that are more adaptive to the era of globalized communication