KGKumar GauravStaff Data Architect

American Megatrends · 2014 - 2017

Master Data & Release Platform

Centralised scattered data into a master data system with regional replication that respected PII and data laws, and built AMISVN, a Python version-control layer on Subversion with release automation.

The problem

Engineering data lived in many places, regions needed local copies under different data rules, and release work depended on manual steps and email chains.

Key decisions

  • Migrate first, then master: one trusted record before anything is replicated.
  • Classify PII before it crosses a region, and replicate only what the law allows.
  • Build version control and release tooling in Python on top of Subversion, instead of replacing what teams already used.
  • Automate submissions and notifications so releases stop depending on email chains.

Outcome

One governed source of master data, replicated safely across regions, and a release process engineering teams relied on. Recognised with the Team Excellence Award in 2015.

PythonMDMReplicationSubversionAutomation

Architecture flow

  1. Scattered source systems
  2. Migration into a central store
  3. Master data management layer
  4. Regional replication with PII controls
  5. AMISVN and release automation
  6. Engineering teams across regions

More architecture work

Book a conversation

Enterprise Data Foundation

A reusable data foundation for risk, products, analytics and AI, not one-off pipelines. Ingestion, medallion warehouse layers, domain marts and operational distribution from a single architectural spine.

Read case study

Global Risk Data Platform

Large-scale company risk-data architecture that ingests, enriches, normalizes and distributes company-level intelligence across countries. Built as a reusable risk pool for underwriting, products, analytics and intelligent systems, and today the data foundation beneath Cowbell's OMNI AI agents.

Read case study

GlueFlux

Metadata-driven processing framework. Pipelines defined in YAML so teams onboard Spark, Python and dbt workloads without rebuilding orchestration, deployment and operational patterns every time.

Read case study