In this demo project, a full-stack machine learning (ML) pipeline is built to forecast time series data and detect model drift in real-time. The pipeline uses PyTorch-based LSTM models for forecasting, InfluxDB 3 for storing time series data and model metrics, and Hugging Face Hub for cloud-based model storage and versioning. The demo showcases how to monitor and adapt ML models as data evolves over time, enabling real-time forecasting and reducing the risk of poor predictions due to model drift. By leveraging the InfluxDB 3 Python Processing Engine, the project can be easily scaled up or down depending on the needs of the system, making it an ideal solution for industries such as industrial IoT, finance, and energy usage monitoring. The demo also highlights the importance of continuous monitoring and adaptation in ML models to ensure reliable predictions.