This study analyzes the phenomenon of code mixing in TikTok video content from the account @salshabilazzahraa, better known as the “Jajan Bareng Lala” account, using a sociolinguistic approach. The purpose of this study is to explore and identify the types of code mixing using Muysken's (2000) theory, namely Insertion, Alternation, and Congruent Lexicalization. Additionally, it classifies their functions according to Appel & Muysken (1987), which include Referential, Expressive, Directive, Metalinguistic, Phatic, and Poetic functions. Using a qualitative research method, data was collected through transcription, classification, and analysis of selected videos. The results of the study indicate that the Insertion type is the most dominant type of code mixing. This indicates that the insertion of foreign words or phrases into the structure of the Indonesian language can be done easily without altering the grammatical structure. This strategy is considered more interesting and in line with the current communication style that is more understandable to many people. Congruent Lexicalization is the second most common type, while Alternation is only found in a limited number of cases in this study. Meanwhile, in terms of function, the referential function is the most dominant. This shows that speakers use code mixing to convey meaning more accurately, efficiently, and contextually. Additionally, the expressive function is also frequently found in this study, particularly for expressing personal opinions, emotions, and self-identity. Meanwhile, the directive function is very rarely found, and the metalinguistic, phatic, and poetic functions are not found at all, as the content in this study is descriptive rather than interactive. This study concludes that code mixing plays an important role in TikTok video content as a means of reinforcing meaning, expression, self-identity, and building closeness with viewers.