We process CI/CD pipelines mostly in the background. We use Sidekiq and schedule asynchronous workers. It is now more and more difficult to avoid race conditions and scale this solution. Event driven architecture for pipeline processing might help with that. This issue is created to discuss concepts of even driven architecture / event sourcing and consider creating a framework which we would be able to use to scale our CI/CD solutions. ### Proposal Build [saga-like process management framework](https://www.cs.cornell.edu/andru/cs711/2002fa/reading/sagas.pdf) to manage domain CI/CD events to improve performance and reliability. /cc @ayufan @pcarranza