MSO Cloud · Documentation

Architecture

Source: docs/architecture/README.md Updated 2026-09-21
On this page

How MSO Cloud is built: the tenant model, the data pipeline that turns ingest into numbers a dashboard can read, the decision engine on top of it, and the rules that keep a number honest.

Audience: engineers and architects integrating with, extending, or reviewing the platform. Read first: Platform Overview for the shape, then System Architecture for the mechanism.

The system#

  • System context — the platform as one box: who uses it, what it reads, what the boundary guarantees.
  • Platform Overview — the tenant model and how an org is composed from plugins and pillars.
  • System Architecture — the monorepo, the two-tier org/brand tenancy model, and the ingest-to-serve pipeline.
  • Extension points for a new customer — every seam that carries a customer-specific answer as data, and the ones that still need code.
  • Data Foundation — the six pillars and three data planes over a layered warehouse.
  • Technology — the stack choices and the constraints behind them.
  • Cloud Architecture — the containerized deployment topology.

Decisions and measurement#

Engineering#

Reference#

Decisions on record#

Every architectural decision is recorded as an ADR under adr/, grouped by theme: platform and BI, composability and UI, connectors and ingest, tenancy and security, metrics and reconciliation, the decision engine and AI, content intelligence, notifications, and vertical/dataset architecture.