Hybrid, AI-augmented migration toolkit combining bulk transpilation (Lakebridge), intelligent gap-filling (AI Coding Assistant + custom skills), and workspace-aware deployment (Databricks AI Dev Kit).
| Artifact | Purpose |
|---|---|
informatica-to-databricks-migration-plan.md |
Complete phase-by-phase migration plan with tool stack, risk register, and success criteria |
Skill Tree/ |
46 AI-ready skills that teach an AI assistant to read Informatica XML and generate Databricks equivalents |
informatica-to-databricks-migration-plan.pdf |
PDF rendering of the migration plan for sharing |
Stack: Lakebridge · AI Coding Assistant · Databricks AI Dev Kit · Custom Informatica Skills
┌─────────────────────────────────────────────────────────────┐
│ INFORMATICA POWERCENTER ESTATE │
│ (Mappings, Workflows, Sessions, BDM, DEI, Parameters) │
└──────────────────────┬──────────────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────────────────┐
│ LAYER 0: UNDERSTAND (AI Assistant + Informatica Skills) │
│ - Read exported repository XML as plain text files │
│ - Explain business logic per mapping/transformation │
│ - Identify dependencies, edge cases, anti-patterns │
└──────────────────────┬──────────────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────────────────┐
│ LAYER 1: CONVERT (Lakebridge + BladeBridge) │
│ - Profile & assess estate complexity │
│ - Bulk transpile 70-80% of standard mappings │
│ - Generate complexity report & reconciler baseline │
└──────────────────────┬──────────────────────────────────────┘
│ (unconverted artifacts + error logs)
▼
┌─────────────────────────────────────────────────────────────┐
│ LAYER 2: FILL (AI Assistant + AI Dev Kit + Skills) │
│ - Read unconverted mapping XML manually via file access │
│ - Rewrite complex workflows as Databricks Workflows/Airflow│
│ - Convert BDM/DEI gaps to PySpark / Structured Streaming │
│ - Translate proprietary functions to Spark SQL UDFs │
│ - Optimize for Delta Lake (Z-ORDER, liquid clustering) │
│ - Validate against live workspace via AI Dev Kit tools │
└──────────────────────┬──────────────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────────────────┐
│ LAYER 3: HARDEN (AI Dev Kit + DABs) │
│ - Bundle as Databricks Asset Bundles (dev/prod) │
│ - Enforce Unity Catalog governance │
│ - Deploy via CI/CD (GitHub Actions / Azure DevOps) │
│ - Scaffold post-migration apps (Genie, AI/BI) │
└──────────────────────┬──────────────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────────────────┐
│ DATABRICKS LAKEHOUSE + UNITY CATALOG │
│ (Delta Live Tables, Workflows, SQL Warehouses) │
└─────────────────────────────────────────────────────────────┘
.
├── README.md # This file
├── informatica-to-databricks-migration-plan.md # Full migration plan
├── informatica-to-databricks-migration-plan.pdf # PDF version
└── Skill Tree/
├── README.md # Skill tree documentation
└── .claude/skills/informatica/ # 46 AI skills
├── informatica-transformations/ # 19 transformation types
├── informatica-workflows/ # 5 workflow categories
├── informatica-repository/ # XML schema, metadata, lineage
├── informatica-functions/ # 6 function categories
└── informatica-to-databricks-mapping/ # Rosetta Stone mappings
| Category | Skills | What They Cover |
|---|---|---|
| Transformations | 19 | Source Qualifier, Expression, Lookup (connected/unconnected, static/dynamic cache), Joiner, Aggregator, Router, Filter, Sorter, Rank, Sequence Generator, Update Strategy, Transaction Control, Normalizer, Stored Procedure, Data Masking, Custom Transformation, External Procedure, Union, XML Source Qualifier |
| Workflows | 5 | Workflow design, session configuration, scheduling, parameter files, error handling |
| Repository | 3 | XML schema parsing, metadata objects, column-level lineage tracing |
| Functions | 6 | Date, string, numeric, conversion, conditional, and aggregate functions |
| Mapping Guides | 5 | Transformation mappings, function mappings, workflow orchestration, performance patterns, anti-patterns |
Start with informatica-to-databricks-migration-plan.md for the full strategy, tool installation steps, phase-by-phase execution, and risk register.
These skills follow the Databricks AI Dev Kit format and work with Claude Code:
# Project-scoped (recommended)
cp -r "Skill Tree/.claude/skills/informatica" ./.claude/skills/
# Or global installation
mkdir -p ~/.claude/skills/
cp -r "Skill Tree/.claude/skills/informatica" ~/.claude/skills/Once installed, ask your AI assistant questions like:
- "Explain how this unconnected lookup works in the attached Informatica XML"
- "Migrate this Source Qualifier SQL override to Spark SQL"
- "What's the Databricks equivalent of this IIF + ISNULL expression?"
- "Parse this repository XML and trace column lineage"
The AI assistant will automatically load the relevant skills and cross-reference them.
| Metric | Target |
|---|---|
| Mappings auto-converted by Lakebridge | ≥ 70% |
| Mappings requiring AI assistant intervention | ≤ 30% |
| Data reconciliation pass rate | 100% (row-level + aggregate) |
| Workflow orchestration migrated | 100% (redesigned, not 1:1) |
| Informatica decommissioned | Domain-by-domain, 100% final |
| Post-migration query performance | ≤ 120% of Informatica baseline |
| Unity Catalog governance coverage | 100% of migrated tables |
- Confirm BDM/DEI scope; decide if commercial accelerator needed
- Export PowerCenter repository XML + download all PDF guides
- Install Databricks AI Dev Kit in migration project directory
- Install the Informatica skills (this repo)
- Run Lakebridge assessment against exported repository
- Establish complexity matrix and gap registry
- Begin bulk transpilation (Phase 1)
See the full migration plan for detailed execution steps.
Skills are derived from official Informatica PowerCenter 10.4.0 documentation:
PC_1040_TransformationGuide_en.pdfPC_1040_(XML)Guide_en.pdf
This project is licensed under the Apache License 2.0.
The migration plan, skill structure, and Databricks/Spark mappings are original work released under Apache-2.0.
This repository references third-party tools that are separate projects with their own licenses:
- Databricks Lakebridge — Databricks Labs
- Databricks AI Dev Kit — Databricks Solutions
The skills are derived from Informatica's publicly available documentation. Informatica, PowerCenter, and related marks are trademarks of Informatica LLC. Databricks and related marks are trademarks of Databricks, Inc. This project is not affiliated with or endorsed by Informatica LLC or Databricks, Inc.
Code examples and mappings in this repository are illustrative. Validate all generated code in your own environment before production use.