Architect

Work With Data Engineers To Translate Logical Models Into

Bronze event tables

Data Modeler

Primary purpose

The Data Modeler defines how the organization represents its business concepts consistently across events, Bronze, Silver, Gold, semantic models, and ontology.

The Data Modeler Answers Questions Such As

• What is a member-month?
• How is a delegated population represented?
• How do contract, risk arrangement, product, and population relate?
• How do we preserve restatements and historical versions?
• What is the grain of each data product?
• Which entity owns each business key?

Key responsibilities

• Enterprise and conceptual modelling
• Logical data modeling
• Define entities, attributes, relationships, cardinality, and business keys.
• Physical modeling
• Work With Data Engineers To Translate Logical Models Into Bronze event tables
• Silver conformed entities
• Gold analytical data products
• Lakehouse tables
• Relational operational stores
• Star schemas
• Aggregate tables
• Cube.dev models
• Graph or ontology representations
• Temporal and historical modeling
• This is especially important for the program.

The Data Modeler Must Differentiate

The Data Modeler works with the Cube.dev and ontology teams, but does not replace them.

The Modeler Ensures That

• Physical data structures support governed measures.
• Cube.dev dimensions and joins reflect valid business relationships.
• Ontology concepts map to identifiable data entities.
• Data grains do not produce double-counting.
• Business hierarchies are represented consistently.
• Semantic definitions remain connected to source data and lineage.

Required Skills

• Conceptual, logical, and physical data modeling
• Relational and dimensional modeling
• Event and canonical-data modeling
• Effective-dated and temporal modeling
• Bitemporal modeling
• Business-key and surrogate-key design
• Master and reference data
• Data normalization and denormalization
• Lakehouse and warehouse modeling
• Metadata and lineage
• Model governance and versioning
• Strong SQL knowledge
• Modeling tools such as ER/Studio, erwin, Sparx, Archi, or equivalent

Preferred Domain Knowledge

• Healthcare claims and encounter models
• Membership and eligibility
• Provider and organization hierarchies
• Capitation and delegated-risk arrangements
• Provider contracts and settlements
• Finance and accounting data
• Journals, subledgers, and GL
• Actuarial and reserve data
• Semantic layers and knowledge graphs