Centralized Disaster Response and Inventory Management System that leverages AI and Google Cloud Technologies to predict disasters, optimize resource management, and provide real-time coordination.

Compare the performance of four deep learning models—CNN, MobileNetV3 Large, VGG16, and EfficientNetB4—for the automated detection and classification of skin lesions

A project on building deep learning classifier to classify playing cards

A 5-layer CNN model for facial emotion recognition trained on FER-2013. Achieved 76% validation and 63% test accuracy using data augmentation. Key features include convolutional layers, max pooling, and dropout. Suitable for human-computer interaction applications.