This paper describes the design and implementation of an earthquake prediction and alerting system that combines readily available, inexpensive hardware, such the Raspberry Pi, with machine learning. An accelerometer (MPU), vibration sensor, and BMP sensor (which measures temperature, pressure, and altitude) are built into the system to continuously monitor seismic and environmental conditions. Key seismic characteristics including Magnitude, Root Mean Square (RMS), and Peak Ground Acceleration (PGA) are computed by analyzing real-time sensor data. The accuracy of several machine learning algorithms is then tested and trained using this data to create a comprehensive dataset for earthquake prediction. The most accurate machine learning model is chosen for deployment after it has undergone a thorough evaluation. For real-time predictions, the selected model is implemented on the Raspberry Pi with a Flask server. The system’s capacity to send SMS notifications based on the measured magnitude of earthquakes and automatically trigger alerts through Twilio’s messaging API is one of its key features. By alerting users to impending seismic activity, these alerts help for prompt evacuation or other safety precautions. This approach offers a scalable, effective, and affordable way to improve public safety in seismically vulnerable locations in addition to showcasing the promise of using IoT and machine learning for earthquake forecasting.