LearnAdapt Agentic Studio lets teachers build AI teammates without code

Teachers and researchers often struggle to customize educational AI to local needs because most systems require programming expertise. To address this, Nizam Kadir introduces LearnAdapt Agentic Studio, built on the PedOS 1.1 Lumina runtime environment. The platform allows non-coders to describe a learning interaction in plain English; the system then generates a previewable plugin artifact, performs automated safety checks, and provides a submission workflow for review. Once approved, plugins are deployed into a central directory where teachers can install them directly into their classrooms.

What sets LearnAdapt apart is its focus on governance and privacy. Telemetry data collection is strictly gated—only authenticated users running approved plugins can contribute usage metrics, ensuring compliance with educational data regulations. This framework transforms AI from a static tool into a dynamic, teacher-built teammate. Accepted at the 27th International Conference on Artificial Intelligence in Education, the project demonstrates a complete lifecycle from natural language prompt to governed, evidence-capturing educational plugin.