In this liveProject, you’ll help takeout delivery company FTGO Inc. take their first steps in moving to a microservices-based architecture. You’ll begin by creating Kubernetes manifests that enable you to deploy your services and their required infrastructures, such as databases, within a cluster environment. You'll use the package manager Helm to seamlessly install and manage infrastructure services like Postgres and Apache Kafka and use kind (Kubernetes in Docker) to create a local Kubernetes cluster that runs on your machine. Finally, you’ll see how GitHub Actions can automate your testing process.

Takeout company FTGO has a problem with its Kubernetes-based application—its services are still deployed using vanilla Kubernetes manifests. This makes it hard to tailor services to environments and automatically declare a service’s dependencies. You’ve stepped in to help them—by utilizing Helm, a Kubernetes package manager. You’ll get hands-on experience creating Helm charts for services, configuring deployment pipelines to seamlessly publish them to Helm repositories, and mastering the art of deploying and upgrading services packaged as Helm charts.

Takeout e-commerce company FTGO currently manages their Kubernetes microservices deployments from the command line. So there’s no record of who has deployed what, and it’s very difficult to reproduce past states. Your goal in this liveProject is to help them with a GitOps-based continuous deployment pipeline for their platform. You’ll utilize Flux to establish and build your pipeline and configure the Flagger tool to use the traffic management features of the Linkerd service mesh.

FGTO is almost ready to deploy their Kubernetes application—they just need you to help them create a Kubernetes cluster for their production environment. The simplest and most reliable way to do this is to use a managed Kubernetes service, and for this liveProject, you’ll use AWS EKS. You’ll work with various tools, including Terraform and Crossplane, to create and configure the cluster. Terraform helps you configure your cluster with just a small amount of code, while Crossplane ensures you can provision the service and its infrastructure using only Kubernetes manifests. Once the application is deployed, you’ll work to encrypt secrets in Git using SOPS and AWS KMS and promote changes from development to production on EKS. When you’re done, FGTO will have a complete working Kubernetes deployment!