The Untold team has worked with many different backend stacks. As we have done for years, we intend to continue to use the best tool for the job; and therefore, we don't have a one-size-fits all deployment pipeline.
We've had success using Capistrano for Ruby and miscellaneous filesystem deployments, and Fabric for basic Django and Python deployments. For PHP applications, we're big fans of the Laravel ecosystem of tools, and tend to launch all our development, and a good number of our production, sites, and applications using Laravel Forge.
For more complicated deployments, and for projects requiring more sophisticated CI/CD pipelines, we've historically looked to Jenkins to handle automation. Philosophically, we believe in testing every single pull request against both development (for finding collisions with other in-progress features) and current production instances of all our apps.
For Drupal, it can often make sense to lean on larger platform-as-a-service companies with great tooling and support. Our favorites are Pantheon and Amazee.io, though we have a ton of experience building, maintaining, and launching both on Acquia's cloud, and on our own custom AWS instances, containers, and systems.
Regardless of final destination, it almost always (except when organizations come with their own processes and environments) makes sense to spin up development and staging/QA servers and services on AWS. This makes active development dramatically more efficient, and helps developers acclimate to and understand the nuances of actually running a given application in a live-like environment.
Web developers can't simply be programmers any more. It's absolutely critical to understand the specific contexts in which your code will run.
