Playbook
0 / 8 complete0%
  1. 01
  2. 02
  3. 03
  4. 04
  5. 05
  6. 06
  7. 07
  8. 08
Step 03 of 8 4-6 weeks· advanced

Step 3: Target Architecture and Framework

Stand up the Azure-side architecture and the metadata-driven pipeline framework before migrating any specific pipeline. Reuses Phases 2-5 of the Azure Data Platform Modernization playbook.

What you're doing in this step

Make the Fabric vs Synapse vs Hybrid decision, set up the metadata-driven ADF / Fabric pipeline framework, deploy the PySpark transformation framework with shared utility library, integrate the data quality framework. Don't reinvent — use the Azure Data Platform Modernization Playbook's framework setup as your starting point.

Recommended prompts

Use one of these to do the work in your IDE

Open the template to read it in full. Click Copy prompt to grab it (with your stack values pre-filled where they apply) — then paste into Claude Code, Cursor, or wherever you build.

Primary recommendation 1-2 days for initial framework design

ADF / Fabric Pipeline Design Patterns

Design patterns for Azure Data Factory and Microsoft Fabric pipelines: metadata-driven, parameterized, idempotent, and observable.

azureadffabric
View template
Template· Template reference document; 1 day for initial setup

PySpark Transformation Standards

Standards for PySpark transformations in Bronze/Silver/Gold pipelines: idempotency, partitioning, Delta Lake patterns, and code organization.

Use this when: Setting up the PySpark side of the framework specifically

azurepysparkfabricdatabricks
View template
Template· Template 1-2 days for initial setup

Data Quality Test Suite

Generate comprehensive data quality tests for ETL pipelines: schema validation, freshness checks, null/duplicate/range checks, and business invariants.

Use this when: Building the DQ framework as part of the platform setup

azurepysparkfabric
View template
Playbook· Playbook 2-4 quarters

Azure Data Platform Modernization Playbook

Build or modernize a cloud data platform on Azure: ADF/Fabric pipelines, PySpark transformations, Synapse/Fabric warehouse, BI consumption, and data governance.

Use this when: You want a complete framework-build sequence to follow alongside this playbook

azureadffabricpysparksynapse
View playbook
Recommended skills

Drop these into Claude Code for this phase

Skills auto-trigger on the right kind of request. Install once; they apply to every prompt that fits.

Skill· Skill 5 min setup

Migration Planner Skill

Flagship migration skill that walks Claude Code through audit → strategy → slicing → cutover for any legacy system migration.

claude-code
View skill
Skill· Skill 5 min setup

Compatibility Layer Generator Skill

Claude Code skill that generates adapters, shims, and proxies for incremental coexistence between legacy and new systems.

claude-codedotnet
View skill
Recommended MCP configs

Wire these tools into Claude Code first

MCP servers give Claude Code direct access to external systems (Jira, browsers, databases). Configure once.

MCP config· MCP config 10 min setup

Azure DevOps MCP for Evoke

Pre-configured Azure DevOps MCP server for Claude Code — work items, repos, PRs, and pipelines from chat.

claude-codemcp
View config
MCP config· MCP config 5 min setup

Filesystem MCP for Evoke

Pre-configured filesystem MCP server for Claude Code — safe, scoped read/write access to project files.

claude-codemcp
View config
When you're done

Verify these in your own work before moving on

This is a checklist for you to mentally tick off in your repo and IDE — the site doesn't track it, you do.

  • Target architecture documented (Fabric / Synapse / Hybrid choice)
  • Pipeline framework operational in dev
  • PySpark framework with shared utility library
  • DQ framework integrated into pipelines
  • First "hello world" pipeline migrated end-to-end (proves the framework)
Common pitfalls

What goes wrong at this step

  • Skipping the framework — hand-coding 350 ADF pipelines is the path to madness
  • Choosing tools without piloting — validate the architecture with one real pipeline before declaring done
← Previous step

Command Palette

Search for a command to run...