The repo showcases how we can automate the provisioning and decommissioning of endpoints for hosting custom comprehend models, to optimize the cost. - View it on GitHub