How we built a DDD inspired, model-driven event-sourced platform (and lived to tell about it)

Early 2015 at Luminis we started on a project with a client in the education publishing space who envisioned a next generation online educational product that would provide educational institutions with an integrated solution for educational content catalogs, learning management, student tracking, curriculum planning, classroom scheduling and much more.

Recognizing the value of Domain Driven Design we thought this would be a prefect use-case, as it was a highly complex business domain with a lot of uncertainties and a solution that would need to be able to grow and evolve over time. It also quickly became clear to us that the need for flexibility and the sheer amount of projected (no pun intended) users and data called for an event-sourcing based CQRS architecture.

That is how it started and now, some 8 years later, we would like to share what we have learned along the way: the challenges of event-sourcing in a DDD context, how we managed to deal with ever evolving domains, explore why immutable journals are cool (until they are not), how we addressed the challenges of keeping things consistent and performant and scalable at the same time.