Continuing from the previous episode, we look at speeding up the build times with a remote builder, continued warning about accessories, implementing background jobs, setting up the infrastructure for background workers and diving more into the Kamal (MRSK) commands.

For various reasons, we may decide to move our services off of one platform onto another. This could be due to pricing, uptime or other reasons. In this episode, we look at migrating a Ruby on Rails application and database from Heroku to Fly.

Continuing on from the previous episode, we look at creating a Docker Swarm cluster and using Traefik to load balance and route requests to various applications. Using this and CloudFlare proxies, we're able to create and deploy hobby applications in minutes.

As an application grows in popularity and usage you'll need to scale the application to support your new users and their data. One way in which your application may need to scale is on the database level. In this episode, we look at reading from a replica database while writing to a primary database. We also look at structuring our application top support multiple databases; each containing their own schema.