Summary

In this chapter, we have discussed some of the most important concepts when it comes to deploying your ML solutions. In particular, we focused on the concepts of architecture and what tools we could potentially use when deploying solutions to the cloud. We covered some of the most important patterns used in modern ML engineering and how these can be implemented with tools such as containers and AWS Elastic Container Registry and Elastic Container Service, as well as how to create scheduled pipelines in AWS using Managed Workflows for Apache Airflow. We also explored how to hook up the MWAA example with GitHub Actions, so that changes to your code can directly trigger updates of running services, providing a template to use in future CI/CD processes.

In the next chapter, we will look at the question of how to scale up our solutions so that we can deal with large volumes of data and high throughput calculations.