MLflow

MLflow is central to our MLOps practices. It provides a unified interface to track experiments (hyperparameters, metrics), package code for reproducible runs, and manage the model registry. By integrating MLflow into our pipelines, we ensure that every deployed model can be traced back to the exact code and data version used to train it.

MLflow is an open-source platform designed to manage the machine learning lifecycle, including experimentation, reproducibility, and deployment. It provides a comprehensive set of tools for tracking experiments, packaging code into reproducible runs, and sharing and deploying models across various environments.

It can be seamlessly integrated with nearly no code changes into models using the Lightning framework, but it also supports other frameworks like TensorFlow, Scikit-learn, and XGBoost.