Data Mesh for Global ShippingData Mesh for Global Shipping
The Challenge
A global bulk-liquid shipping leader struggled with siloed data across fleet management, logistics, and customer service. Centralized architecture could not keep up with demand for real-time insight. Domain teams lacked ownership of their own data, and multi-client datasets on shared infrastructure raised separation and compliance challenges.
The Solution
Coditas built a Data Mesh on Azure Databricks with domain-owned, discoverable data products. The architecture used per-tenant medallion layers in separate schemas or storage paths, Kimball dimensional modeling, automated ingestion via Databricks Workflows and Azure DevOps, and decentralized governance with Unity Catalog ABAC, lineage, and tenant-configurable retention.
The Outcome
Domain teams gained ownership of their data products and self-serve insight, reducing reliance on central IT. The solution enabled near-real-time decisioning, consistent data quality across domains, and a scalable foundation for advanced analytics.
Engineered With: Azure Databricks · Apache Spark · Spark SQL · Unity Catalog ABAC · Power BI · Azure DevOps
Our collaboration with Coditas grew from one specialist to more than 15 professionals based on the quality of the work and collaboration. The team supported us from designing a new data lakehouse architecture through its implementation in Databricks, while also contributing across BI, data engineering, and data science. What stood out was their willingness to understand our industry before implementing solutions.
Build an AI Strategy Your Teams Can Act On
Ashutosh Zatke
Director - Growth Strategy
When you win, we win.
Our Offices
