Back to projects
Data Engineering

Danish Company: Unified Data Platform

A metadata-driven data platform for consolidating operational sources into analytics-ready workflows.

Role
Data platform engineering
Focus
Data Engineering
Stack
Azure Data Factory, Databricks, Azure SQL

Context

The shape of the problem

The project was about creating a reliable foundation rather than a flashy front end. Multiple data sources needed to move through a coherent platform where quality, lineage, and update patterns could be reasoned about.

Build

What I actually built

I helped shape Azure Data Factory and Databricks workflows around metadata-driven ETL, data quality checks, and maintainable orchestration. The outcome was a platform pattern that could scale with additional sources instead of becoming a one-off pipeline.

Engineering

Key decisions

01

Integrating heterogeneous data sources with varying schemas and update frequencies

Developed a metadata-driven ETL architecture with Azure Data Factory

02

Ensuring data quality and consistency across the platform

Implemented advanced data quality rules and monitoring

03

Building a solution that scales with growing data volumes

Created optimized Databricks workflows for large-scale data processing

04

Creating a flexible metadata model to adapt to changing business requirements

Built a self-service data catalog for business users

Outcome

What changed

  1. 01Improved data quality metrics across all integrated sources
  2. 02Enabled new analytical capabilities for client services teams

Stack

Tools and systems

Azure Data FactoryDatabricksAzure SQLAzure Blob StorageAzure Data LakePythonDevOpsAzure Pipelines

I build delightful systems.

© 2026 Kristian Ernst.