Chapter 5: Continuous Deployment of a Production ML Model

In Chapter 4, Continuous Integration and Continuous Delivery (CI/CD) for Machine Learning, we were introduced to the concept of continuous integration, and continuous deployment, as a means of bridging the gap between ML model development and ML model deployment. We were also introduced to the AWS CDK, as a way to further close this gap, by bringing the different artifacts that software engineers and ML practitioners develop into a single cloud-native application. Thus, allowing us to codify a CI/CD pipeline that automates the entirety of the ML process. Closing this gap, and helping to facilitate this inter-team synergy, is one of the core design philosophies behind why AWS originally created the CDK.

Note

For more information on the AWS CDK philosophy, you can read the best practices for developing cloud applications in the AWS CDK blog post (https://aws.amazon.com/blogs/devops/best-practices-for-developing-cloud-applications...