MLflow-instrumented XGBoost training with a hyperparameter sweep, a model registry, and inference against the registered model. This replaces project 4’s manual folders with the tool teams actually use — now every run is comparable, every model is registered, and nothing depends on you remembering which folder had the best result.
MLflow-instrumented training with a hyperparameter sweep and a registry is the concrete, toolable skill most ML Engineer job descriptions list by name.