Whenever you're ready, walk me through how you'd approach this.
What is your experience with model evaluation metrics?
Explain which metrics you have used, how you selected them for different machine learning tasks, and how they informed model comparison or deployment decisions. Include an example from your experience, covering the problem context, relevant metrics, trade-offs, and any error analysis or validation techniques you applied.