Supralog built an online incremental machine learning pipeline with InfluxDB OSS for capacity planning, using Kapacitor, Python, and InfluxDB. They used Kapacitor to monitor prediction residuals and trigger alerts, which would retrain the model if it drifted. The pipeline consists of three components: trend, seasonality, and residuals, each modeled by a different machine learning approach (linear regression for trend, LSTM for seasonality, and another LSTM for residuals). The pipeline uses InfluxDB's TICK Stack to store and query data, and Kapacitor to alert on model drift. The authors discuss the advantages of using InfluxDB and Kapacitor for this use case, including their ability to handle large amounts of data and provide real-time insights. They also highlight the importance of online machine learning and incremental training, which allows the model to adapt to changing conditions without requiring a complete retraining.