Vitiligo is characterized as a skin condition in which there are areas devoid of pigmentation of the skin which results in white patches to be visible. Early diagnosis is essential for treatment to begin and an effective management plan to be initiated. In this project, we designed a Smart Vitiligo Diagnosis System with Raspberry Pi that would allow the user to detect areas on their skin that might be presenting their vitiligo symptoms through skin images. The system will run a trained AI model trained from a dataset of vitiligo and normal skin images. It will utilize a Convolution Neural Network (CNN) to perform feature extraction and classifying the image. The trained model is run on a Raspberry Pi with a Pi Cam 2 (8MP) to capture the images and a display to view the results. The device will be designed so that the user can scan a patch of their skin and get instant data on whether or not vitiligo has occurred. Overall, this project illustrates that machine learning can be leveraged to detect vitiligo in both an accessible and an efficient manner.