Running computer vision on microcontrollers is becoming practical thanks to MCUs that integrate NPUs, enabling low-latency inference with tight power and cost budgets. In this session, Roboflow Co-Founder and CEO Joseph Nelson will walk through an end-to-end workflow for building a custom vision model using the Roboflow platform—covering data preparation, training and export for embedded deployment. Then, Kwabena Agyeman (Co-Founder and President, OpenMV) will show how to deploy the model onto OpenMV’s NPU-enabled low-power smart camera modules, such as the AE3 and N6, and demonstrate it running live on-device. Attendees will leave with a clear, repeatable process for taking a vision idea from dataset to working embedded demo, plus guidance on the kinds of smart camera applications this process unlocks at the edge.