IDRAK prototype 1

IDRAK Prototype 1 was born from a simple frustration: today’s AI is powerful, but largely confined to screens, keyboards, and fragmented apps, far from where real work and learning actually happen. We asked what it would look like if intelligence could live directly in a person’s physical environment, see what they see, and guide them in real time. This became especially compelling in contexts where timing and focus matter deeply, such as hands-on skill acquisition, safety-critical tasks, and Qur’an memorization.

To explore this, we built an early MVP of a human-centered physical AI engine. The system ingests live visual input, builds short-term temporal context from video frames, and routes that perception to an AI reasoning layer to infer user intent and task state. It then returns structured, real-time guidance through visual overlays like bounding boxes, arrows, captions, and safety cues, paired with optional audio instructions. For the hackathon, we focused on validating the core engine rather than a full wearable device, using a laptop webcam and a web-based interface to simulate a smart-glasses experience. The result demonstrated that responsive, grounded assistance in the physical world is achievable with current technology.

What we learned is that the real opportunity is not a single application, but a platform for embodied AI that can support many workflows while keeping humans at the center. As this space emerges, we believe it is important that values-driven builders are present early, helping shape how this technology develops and who it ultimately serves. For us, that includes building responsibly and ensuring communities often left out of frontier technology, including Muslims, have a role in steering its direction. The same engine naturally extends to other high-impact domains where real-time, contextual guidance matters, including lab protocol training, surgical skills education, and many more.

Next, we are focused on translating the engine to wearable hardware, expanding priority use cases, and growing a team aligned with building physical AI with purpose.

**Note: The quran memorization video in the 'try it out' links is not directly related to this project, however, it represents a use case of what we are building.