Shippable is quite different from the competition - that is the first thing that came to my mind while trying to use it. They have innovated a lot to woo both single developers and enterprises alike. I like a few things about it: A. Multi-language support with docker focus - Shippable is not a docker-only nor a docker-unaware platform. It detects the primary programming language of the project very swiftly, and if the project is docker-based, it picks up the docker settings too and spins up a container for the deployment process. It is quite easy to mistakenly assume that Shippable is only for docker projects, but it is not - it just integrates very well with them. Language support is quite good with python, node.js and others on the menu. B. BYOH (Bring your own host) approach - This is where the marriage between enterprise and single users come into play - If you don't like the default VMs provided by Shippable as your testing and building platform, you can swap them out for any Azure, DigitalOcean, Google App Engine, Google Compute Engine or Heroku instance for custom experience. You can also use your on-premises architecture along with the hosted ones to get the best of both worlds. Developers want to use the popular services as VM provider and enterprises want to use their own machines for the best experience. C. Debugging - Full access to console output for every test/build and SSH access into those machines really help in debugging when the tests fail for some unknown reason. D. Auto-configuration from popular test config files - While setting things up, I just activated build for one project without configuring anything for Shippable. This project already had Travis and Circle CI set up for proper testing. Now, I reasonably expected this build to fail with something like "No tests found", but to my surprise it recognised one of those config files and ran all the test commands successfully. Although the test results were not collected, it was overall a pleasant surprise for me.

Shippable is one of the very few CI servers which support only Docker-based or containerised testing/deployment. So, if you are not in that target demographic, you're out of luck. But if you do use Docker, Shippable packs quite a punch and brings fresh approach to the process. Its not a whole lot customisable, but covers a lot of ground with one-click integration with external services. Things that I like about Shippable ============================= 1. Visual Continuous Delivery Pipelines - Usual method for setting up the flow in most CI servers is to write bash code for all the parts of the process in a single file, but Shippable is totally the opposite. You have a nice timeline-like UI where you can activate different service cartridges like databases, programming languages, notification services, AWS deployment, etc. The whole flow is divided into different parts like CI, CD and notification. 2. Huge range of deployment integrations - Supports direct integration with AWS Elastic Beanstalk (EB), Heroku, Openshift, DigitalOcean and more. Because everything is specified in the pipeline visually, they need to provide deployment adapters for many services and they do a very good job at it. 3. Supports many databases and languages within container - Although docker is nice for running everything separately, its success depends on how modular the images can be made. If a lot of things need to be installed at container startup, it is going to be as slow as traditional VMs. So, most databases and languages are modularised into cartridges you can activate externally. 4. Very affordable pricing even for large organisations - The basic account is free, but parallel testing for many containers or multiple deployments from the same pipeline costs $10 for each extra item. Even after 20 of each, the total price comes to about $400 which is ultra cheap considering the volume of such organisations. 5. Separate pipeline for separate branches in code repositories are supported. 6. SSH login for test debugging is very useful when the bugs are not reproducible in local machines.

Integration with third party cloud platforms like Amazon EC2 Container Service (ECS), Google Container Engine (GKE), or Red Hat OpenShift 3. This allows you to easily build and ship