Industrial digital twins have been around for a long time. Industrial metaverse gives a whole new meaning for digital twins as we can start blending the virtual world with the real world through AR. This could be used, for example, in industrial maintenance helping the experts to see more information about the machines or objects or specify the malfunctioning parts of the machine in real-time. In order to make this reality, overlaying the 3D models on top of their real-world counterparts is a challenge. In this project we aim at developing a system that recognizes objects without markers (such as QR codes) using visual features, sensor data, and minimal tagging. The system should match what the camera sees with the 3D model fetched from the database. How might we make a simple demonstration which recognizes the position and angle of the machine and overlays the 3D model in realistic way on top of it? If we had real-time product and operational data, what could we visualize on the 3D model? We are aiming at using simple camera tech and MCP server to take full advantage of AI to fetch the information from the databases. If you are interested in cutting-edge industrial metaverse, this project is for you!