This study aims to develop an IoT-based crib for infants that incorporates AI technology. The system is designed to detect baby crying, the baby’s absence from the crib, the presence of a stranger around the crib, and when the baby is lying face down. An alert message is then sent to the phone of the child’s parent(s), caregiver, or both. The proposed system uses an ESP32-CAM with an I2S microphone (model NMP441) to collect sounds, pictures, and clips, which are then processed and trained using Edge Impulse, ArcFace, and YOLOv5. The system achieved 97% accuracy in detecting baby crying in real-time scenarios, 93.67% accuracy in detecting when the baby is lying face down, 93.5% accuracy in detecting the presence of a stranger around the crib, and 98.8% accuracy in detecting when the baby is missing from the crib. The proposed solution provides a low-cost and efficient method for real-time baby monitoring and ensures the infant’s safety.