The integration of artificial intelligence (AI) into secondary education is rapidly accelerating, yet its use in computer science classrooms remains underexplored. This research investigates how Retrieval-Augmented Generation (RAG) based Large Language Models (LLMs) can serve as supplementary tools for secondary computer science (CS) educators. By focusing on secondary teachers’ needs, this research addresses a critical gap in recent literature as most studies on generative AI (Gen AI) in education have been conducted in higher education or in subject areas outside of secondary computer science.

The research employs a mixed-methods design across three sequential studies: an exploratory landscape study, a classroom intervention, and a synthesis and evaluation. These studies will examine how RAG-based tools can affect artifacts’ alignment to curriculum and standards, quality of instructional materials, and reduction in teacher workload when compared to Commercial-off-the-Shelf (COTS) AI or not using Gen AI at all. The work is guided by Charlotte Danielson’s Domain 1: Planning & Preparation.

Expected contributions include empirical evidence of AI’s impact on secondary computer science education (CS Ed), design principles for effective classroom use of RAG-based systems, and a set of practical guidelines for educators and policymakers. Ultimately, this research aims to make secondary computer science education more equitable, reduce teacher cognitive load, and support workforce readiness by improving the effectiveness of teaching practices in secondary computer science.