Whether you’re building a data warehouse or a lakehouse: we believe in building it on business requirements, by design
When requirements and implementation automatically stay in sync, the warehouse keeps pace.
conflux9 works at the conceptual level, before the logical model
Instead of building a Data Vault model by hand, you define business concepts. You add the relationships, metrics, and KPIs. Everything below is derived from that definition. That includes the logical and physical model, the data transformations, and the orchestration that runs it all. You no longer think in hubs, links, and satellites. That modeling happens for you.
You keep working in SQL as before, only against business concepts instead of Data Vault tables. Your focus stays on the business, from the start and throughout the entire development process.
As a result, much of the development complexity disappears.
From a text-based model to a complete data warehouse
conflux9 is a concept for a system that turns a text-based conceptual layer into the complete technical implementation of a Data Warehouse.
In the conceptual layer, you define business concepts and their relationships. You write the related logic, such as data quality checks, metrics, and KPIs, in SQL that references the business concepts, without needing to know the Data Vault model underneath. The derived data model and transformations fulfill the core requirements of a Data Warehouse: subject-oriented, integrated, time-variant, and non-volatile.
Because the conceptual layer is text-based, it is Git-native: versionable, diffable, and open to code review. When you change the definition, the implementation changes with it.
conflux9 targets modern analytical platforms and fits into your existing toolchain. It is designed to offer native dbt support with full model generation and no manual mapping. Without dbt, it would bring its own SQL transformation engine and scheduler, so no external tooling is required.
It would also include agent skills that assist you in building the conceptual layer. You describe business concepts, their relationships, and their transformations in natural language, and the agent generates it.
Interested in where this is going?
conflux9 is still an idea. We’re working out the concept now, with the aim of laying the groundwork for a tool built on it later. Leave your email to follow how it develops.
Thanks! Please check your inbox to confirm your subscription.
You’re confirmed. Welcome aboard, we’ll keep you posted.
Want to discuss the ideas behind conflux9? We’re happy to talk.
I want to talk to Reinhard about this