Data Architect

- At least 5 years of that experience must include hands-on work with BI tools and reporting platforms, designing data structures to serve BI consumption (Tableau, Power BI, OBIEE or comparable tools). - Hands-on proficiency with Databricks. Must be able to write Spark SQL, and design medallion-layer architectures. - Expert-level SQL skills.

- Hands-on proficiency with Databricks. Must be able to write Spark SQL, and design medallion-layer architectures. - Expert-level SQL skills. - Ability to design repeatable processes, standards, and architectural patterns that enable a team to execute consistently at scale.

Position Overview Halvik is seeking a Data Architect with significant BI experience to lead the technical design and execution of a large-scale BI migration at the U.S. Department of Transportation. This program is migrating thousands of reports and their underlying data structures from Oracle Data Warehouse / OBIEE into a Databricks platform with Tableau as the reporting layer. The Data Architect will design the Databricks gold-layer data models that serve Tableau, architect the Tableau data source (TDS) layer, define and maintain the automated and manual conversion processes that transform legacy report definitions into Tableau workbooks, and resolve the complex structural differences between OBIEE and Tableau (multi-fact decomposition, join strategies, naming authority). A significant part of the conversion pipeline leverages AI/LLM-generated outputs, and the Data Architect will maintain and enhance that pipeline to support the conversion goals.

Halvik is seeking a Data Architect with significant BI experience to lead the technical design and execution of a large-scale BI migration at the U.S. Department of Transportation. This program is migrating thousands of reports and their underlying data structures from Oracle Data Warehouse / OBIEE into a Databricks platform with Tableau as the reporting layer. The Data Architect will design the Databricks gold-layer data models that serve Tableau, architect the Tableau data source (TDS) layer, define and maintain the automated and manual conversion processes that transform legacy report definitions into Tableau workbooks, and resolve the complex structural differences between OBIEE and Tableau (multi-fact decomposition, join strategies, naming authority). A significant part of the conversion pipeline leverages AI/LLM-generated outputs, and the Data Architect will maintain and enhance that pipeline to support the conversion goals. The ideal candidate combines data architecture expertise with substantial hands-on BI experience. The Data Architect must understand how data models are consumed by BI tools, how report logic maps through presentation and logical layers, and how to design data structures that support both automated report generation and ad-hoc analysis at scale.

Core Responsibilities - Maintain the Databricks gold-layer data models that serve as the foundation for all Tableau data sources and reports. Ensure models support both automated report generation and ad-hoc analytical use. - Architect the Tableau data source (TDS) layer, defining how gold-layer tables are surfaced as published data sources, including single-fact and multi-fact structures, join strategies, extract vs. live connection decisions, and performance optimization patterns.

- Maintain the Databricks gold-layer data models that serve as the foundation for all Tableau data sources and reports. Ensure models support both automated report generation and ad-hoc analytical use. - Architect the Tableau data source (TDS) layer, defining how gold-layer tables are surfaced as published data sources, including single-fact and multi-fact structures, join strategies, extract vs. live connection decisions, and performance optimization patterns. - Own the repeatable report conversion process - covering both the LLM-assisted path (AI-generated report scaffolds reviewed and refined by BI developers) and the manual path (structured step-by-step procedures). Define quality gates and validation criteria for each path.

- Own the repeatable report conversion process - covering both the LLM-assisted path (AI-generated report scaffolds reviewed and refined by BI developers) and the manual path (structured step-by-step procedures). Define quality gates and validation criteria for each path. - Resolve the structural differences between OBIEE and Tableau architectures. - Collaborate with Data Engineers on ETL pipeline design, ensuring transformation logic produces gold-layer structures optimized for Tableau consumption. Provide architectural direction on Unity Catalog governance, metadata management, and access control.

- Resolve the structural differences between OBIEE and Tableau architectures. - Collaborate with Data Engineers on ETL pipeline design, ensuring transformation logic produces gold-layer structures optimized for Tableau consumption. Provide architectural direction on Unity Catalog governance, metadata management, and access control. - Design validation frameworks - including source-to-target reconciliation queries, fidelity checks, and automated comparison pipelines - that confirm migrated reports faithfully reproduce legacy business logic.

- 10+ years of experience in data architecture, including designing dimensional models for analytical and reporting environments. - At least 5 years of that experience must include hands-on work with BI tools and reporting platforms, designing data structures to serve BI consumption (Tableau, Power BI, OBIEE or comparable tools). - Hands-on proficiency with Databricks. Must be able to write Spark SQL, and design medallion-layer architectures.

Preferred Expertise - Experience with BI platform migrations (OBIEE-to-Tableau, or similar conversions). - Experience using AI/LLM tools in a professional workflow - particularly for code generation, metadata analysis, or pipeline automation.

- Python proficiency for scripting, automation, or metadata analysis. - Experience with Tableau APIs - Experience working in a federal government IT environment.

- Familiarity with OBIEE RPD architecture (physical, logical, presentation layers). - Python proficiency for scripting, automation, or metadata analysis. - Experience with Tableau APIs