IoT devices produce ever growing amounts of data. Traditional cloud-based approaches for processing data are facing some limitations: bandwidth might become a bottleneck and sensitive data should not leave user devices as stated by data protection regulators such as GDPR. Federated Learning (FL) is a distributed Machine Learning paradigm aiming to collaboratively learn a shared model while considering privacy preservation. Clients do the training process locally with their private data while a central server updates the global model by aggregating local models. In the Computing Continuum context (edge-fog-cloud ecosystem), FL raises several challenges such as supporting very heterogeneous devices and optimizing massively distributed applications. We propose a workflow to better support and optimize FL systems across the Computing Continuum by relying on formal descriptions of the underlying infrastructure, hyperparameter optimization and model retraining in case of performance degradation. We motivate our approach by providing preliminary results using a human activity recognition dataset showing the importance of hyperparameter optimization and model retraining in the FL scenario.