BLAST is a game-based bilingual literacy assessment platform that uncovers students’ strengths, tracks their growth, and equips school and district leaders with the data needed to make informed decisions. Through BLAST, we are developing a corpus of oral assessment responses from bilingual and multilingual children. Each audio recording in the dataset is accompanied by a transcript and metadata that aims to support sociolinguistic and educational research. The planned metadata fields will include: child’s age, grade level, gender, ethnicity, geographical location, home language(s), type of educational program (e.g., dual language, monolingual), language in which each question was asked. Our goal is to support the development of culturally responsive assessment tools and AI models that reflect the strengths and needs of bilingual and multilingual learners in the U.S. and beyond.