Skip to content

Platform Overview

DataMuru is designed to become a productized control layer for enterprise data platform infrastructure. The long-term product thesis is that data platform assembly work should be standardized, versioned, and governed in the same way cloud infrastructure became standardized through infrastructure-as-code.

Today, platform teams often combine manual Databricks UI changes, notebooks, scripts, IAM tickets, spreadsheet-based access reviews, and provider-specific runbooks. DataMuru exists to move that work into one reviewable lifecycle:

declare -> validate -> diagnose -> plan -> apply -> observe -> adopt -> govern

Product direction

DataMuru aims to give platform teams a single control plane model for:

  • Workspace and platform topology
  • Governance taxonomy and policy compilation
  • Access-control baselines
  • Repeatable project bootstrapping
  • Future lifecycle workflows such as import, policy rollout, and compliance reporting

Product pillars

Pillar What it means in the product
Declarative platform intent Teams describe catalogs, schemas, access, and governance in configuration.
Deterministic change review Plans explain what will be created, updated, skipped, or destroyed.
Provider-backed execution Provider adapters perform supported live operations.
Governance-aware modeling Taxonomy, RBAC, and masking are core concepts, not afterthoughts.
Brownfield adoption Existing resources can be discovered, generated into YAML, and adopted deliberately.
Open-core expansion OSS owns shared contracts; Enterprise extends high-scale and regulated workflows.

What DataMuru is not

DataMuru is not intended to be:

  • a replacement for Databricks itself;
  • a generic workflow scheduler;
  • a data transformation framework;
  • a dashboarding product;
  • a hidden wrapper around Terraform;
  • a one-off script generator.

It is a product framework for data infrastructure and governance lifecycle management.

Product principles

  • Python-first: the CLI and automation layer are built around a typed Python core.
  • Declarative-first: configuration expresses desired state, not shell choreography.
  • Governance-first: governance is not a sidecar feature; it influences resource modeling.
  • Provider-ready: platform-specific logic lives behind an implementation contract.
  • Enterprise-credible: documentation, schemas, and change surfaces must be clear enough for global teams and implementation partners.

Current alpha positioning

The current repository is a bootstrap foundation for:

  • A stable contributor baseline
  • A product documentation baseline
  • A future path to richer provider logic and enterprise-grade workflows

This means the alpha should be treated as the beginning of the framework, not the full market promise yet.

Product maturity model

Stage What users should expect
Local evaluation Full config/plan/apply loop against local state.
Workspace smoke test Live Databricks catalog/schema and grant operations on test resources.
Brownfield review Discovery and generated YAML for supported resources.
Enterprise pilot Identity and multi-team operating controls in sandbox environments.
Production rollout Requires stronger state backends, policy maturity, and team controls.

Use the evaluation checklist and enterprise rollout playbook to choose the right maturity stage.