Individualized learning support is a significant challenge for students. Providing individualized tutoring for each student is often not feasible. Embodied virtual tutors could be a promising solution to this problem. These virtual tutors are AI-driven agents that can explain complex concepts, answer questions, monitor learning progress, and provide support to students. In addition, virtual tutors can be personalized and trained with expert knowledge. The project aims to investigate the potential, acceptance, and effectiveness of virtual tutors for individualized student learning. To this end, we are implementing a virtual tutor by leveraging recent advances in artificial intelligence (e.g., LLMs) and mixed reality technology. We will conduct user studies to improve the usability of our prototype. Finally, we will conduct a study on the acceptance of the virtual tutor. In the future, our system could be applied to other teaching environments and scenarios. Translated with DeepL.com (free version)