Tutorials¶
Welcome to the CSA-in-a-Box tutorial series. These tutorials guide you through deploying, configuring, and extending the Cloud-Scale Analytics platform on Azure — from foundational infrastructure through AI-powered analytics and marketplace publishing.
New here?
Start with 01 — Foundation Platform. Every other path depends on it.
Learning path¶
graph TD
START([Start Here]) --> T01[01 - Foundation Platform]
T01 --> BRANCH{Choose Your Path}
BRANCH --> GOV[Path B: Governance]
BRANCH --> GEO[Path C: GeoAnalytics]
BRANCH --> STREAM[Path D: Streaming]
BRANCH --> AI[Path E: AI Analytics]
BRANCH --> MKT[Path F: Marketplace]
GOV --> T02[02 - Governance & Compliance]
GEO --> T03[03 - GeoAnalytics OSS]
T03 --> T04[04 - ArcGIS Integration]
STREAM --> T05[05 - Real-Time Streaming]
AI --> T06[06 - AI Analytics]
T06 --> T07[07 - AI Agents]
T06 --> T08[08 - RAG Pipelines]
T08 --> T09[09 - GraphRAG]
MKT --> T10[10 - Marketplace Publishing]
BRANCH --> DAB[Path G: Data API Builder]
DAB --> T11[11 - Data API Builder]
T11 --> MKT
style START fill:#0078d4,color:#fff
style DAB fill:#00a4ef,color:#fff
style T01 fill:#107c10,color:#fff
style BRANCH fill:#ffb900,color:#000
style GOV fill:#5c2d91,color:#fff
style GEO fill:#008575,color:#fff
style STREAM fill:#d83b01,color:#fff
style AI fill:#0063b1,color:#fff
style MKT fill:#767676,color:#fff All paths begin with 01 — Foundation Platform, which deploys the core Azure Landing Zone, Data Management Landing Zone, and Data Landing Zone infrastructure.
Path A — Foundation (required)¶
-
:material-foundation:{ .lg .middle } 01 — Foundation Platform
3–4 hours. Deploy ALZ, DMLZ, and DLZ with storage, Databricks, Synapse, Data Factory. Run your first dbt pipeline with USDA data through Bronze → Silver → Gold.
Prerequisites: Azure subscription (Contributor+), Azure CLI 2.50+, Bicep CLI, Python 3.11+, Git.
Path B — Governance & compliance¶
-
02 — Governance & Compliance
2–3 hours. Configure Microsoft Purview for data cataloging, deploy Azure Policy guardrails, set up sensitivity labels, and implement row-level security.
Prerequisites: Path A complete, Microsoft Purview access, Microsoft Entra ID P1+.
Path C — GeoAnalytics¶
-
03 — GeoAnalytics OSS
90 min. Deploy PostGIS, process GeoParquet, H3 hexagonal indexing, and Apache Sedona on Databricks for spatial analytics.
Prerequisites: Path A complete.
-
04 — ArcGIS Enterprise (BYOL)
2 hours. Provision Azure infrastructure for ArcGIS Enterprise, configure enterprise geodatabase, and publish feature services.
Prerequisites: Path A + valid Esri ArcGIS Enterprise license (BYOL).
Path D — Real-time streaming¶
-
05 — Real-Time Streaming (Lambda)
90 min. Deploy Lambda architecture with Event Hubs, Stream Analytics, Azure Data Explorer, and Cosmos DB. Build a real-time earthquake monitor.
Prerequisites: Path A complete.
Path E — AI analytics¶
-
06 — AI Analytics with Foundry
90 min. Deploy Azure AI Foundry with GPT-5.4, build a data-aware chatbot, deploy to Container Apps.
-
07 — AI Agents (Semantic Kernel)
90 min. Build single and multi-agent systems with Semantic Kernel, plugins, GroupChatOrchestration, and MCP tools.
-
08 — RAG with AI Search
90 min. Hybrid vector + keyword + semantic reranking search; build a RAG chatbot over your data catalog.
-
09 — GraphRAG Knowledge Graphs
90 min. Build knowledge graphs with Microsoft GraphRAG, Cosmos DB Gremlin, and hybrid graph + vector search.
Prerequisites for Path E: Path A complete, Azure OpenAI access approved. Tutorial 09 also requires Cosmos DB (Gremlin API).
Path F — Marketplace publishing¶
-
10 — Data Marketplace
60 min. Register data products, run quality assessments, manage access requests, and sync with Purview catalog.
Prerequisites: Path A + Cosmos DB deployed.
Path G — Data API Builder & APIM gateway¶
-
11 — Data API Builder
90 min. Deploy Azure SQL + DAB on Container Apps, expose domain data as REST & GraphQL APIs, build a frontend catalog, and integrate with APIM as the unified Data Mesh gateway.
Prerequisites: Path A + Azure SQL + Azure Container Apps.
Quick-start recommendation¶
If you are new to CSA-in-a-Box, follow this order:
- 01 — Foundation Platform (required for all paths).
- 02 — Governance (recommended for production readiness).
- Pick the path that matches your workload: GeoAnalytics, Streaming, AI, Marketplace, or Data API.
Conventions¶
- Each tutorial has a
validate.shscript that verifies successful completion. - Code blocks prefixed with
$are shell commands; those without are expected output. - All resource naming follows the pattern
{prefix}-{service}-{environment}(e.g.,csa-dlz-dev). - Estimated times assume familiarity with Azure CLI and basic cloud concepts.