Machine learning from prototyping to production using Kubeflow

I would like to get help from an expert of Kubeflow so we can save time to deploy it and design the best solution for the continuous training and production inference.