German enterprise analytics platform

Domain: Restaurant & operations (EU) Stack: Python · Spring Boot · Angular · eCharts Since: 2024 production dashboards

Three-layer data platform: Python ingests messy sources, Spring serves governed APIs, Angular + eCharts gives operators drill-down — not one-off BI exports.

Python ETLSpring BootAngularPostgreSQLeCharts

Pipeline

Ingest → stage → serve

POS and finance exports land in staging via nightly Python jobs. Spring exposes versioned REST aggregates. Chart configs stored as data — new metrics without template rewrites.

Why three stacks

Layer separation

Layer	Role	Separate because
Python	Ingest, cleanse, heavy SQL	Fast iteration on source file chaos
Spring	Auth, APIs, business rules	Typed contracts; cacheable aggregates
Angular	Operator UI	Reusable eCharts across business units

Scale & reliability

Tactics

Idempotent batches keyed by source_batch_id
Materialized summaries for sub-second dashboard loads
Feature flags on new metrics — regional opt-in before default

Production outcomes

Results

Operations teams use dashboards daily — not slide-ware
Governed API layer replaced ad-hoc spreadsheet exports
Pipeline extensions ship without front-end rewrites

German buyer programme. Client identity available under NDA.