Designing a data-literate enterprise at scale

Overview

Miridih re-architected its data platform on Databricks, migrating to a serverless environment on AWS with Lakeflow Jobs, Unity Catalog and a medallion architecture, and introduced AI/BI Genie so nontechnical employees can query data and gain insights independently, reducing data processing time by more than 75%.

The challenge

As usage of Miridih's MiriCanvas design service grew rapidly, its legacy architecture relied on disjointed data pipelines and an inefficient computing environment, lacked a robust governance framework, made schema changes difficult, and left key departments unable to make timely, data-driven decisions with organization-wide self-service analytics out of reach.

The solution

Miridih fully re-architected its data platform with Databricks, migrating to a serverless environment on AWS integrated with AWS Glue, implementing Lakeflow Jobs for faster data pipelines, using Unity Catalog with column masking and permission controls for governance, adopting a medallion architecture for data quality, and introducing AI/BI Genie so nontechnical users can query data and gain insights independently.

Reported business value

Miridih reduced the time required to collect and convert customer behavior data from over four hours to under one hour, a performance improvement of more than 75%, and within five months more than 70% of its workforce had adopted Databricks in their daily work, with over 200 dashboards and 180 workflows in use.