Metadata-Driven Data Engineering in Microsoft Fabric
Accelerate your data engineering lifecycle. F³ is a metadata-driven framework based on lightweight JSON configs and reusable artifacts that generates the Medallion architecture in a simple, fast, and structured way.
Implement Microsoft Fabric Right
Microsoft Fabric can be implemented well and bad. If you implement it badly, you will probably suffer it a lot in future development and maintenance.
- •Manual pipeline and notebook authoring for each new table.
- •Hardcoded workspace connections making DEV / TEST / PRO promotions painful.
- •Difficult incremental tracking and manual SCD2 logic prone to bugs.
- •High long-term maintenance costs and unclear data lineage.
- •Reusable generic pipelines driven entirely by lightweight JSON configs.
- •Environment Variable Libraries for clean multi-environment synchronization.
- •Automated incremental time watermarks and native SCD2 dimensions.
- •Automated field and table lineage logging in the control layer.
How the Framework Operates
F³ separates configuration metadata from execution code, generating a robust Medallion architecture (Bronze → Silver → Gold) across Microsoft Fabric.
Control Lakehouse & Warehouse
Config JSON files specify tables, schemas, and load modes. Manages time watermarks for incremental loading and Fabric Variable Libraries for DEV/TEST/PRO environments.
Pipelines & Notebooks
Generic parameterized pipelines (pl_load_to_bronze & pl_load_to_silver) dynamically read metadata to extract, validate, and conform tables.
Bronze, Silver, Gold
Source-isolated Bronze Lakehouses, unified conformed Silver Lakehouse with SCD2 & Surrogate Keys, and business Gold Warehouses.
Workflows and pipelines are controlled by lightweight JSON configuration files instead of manual pipeline authoring for every table.
Automates the generation of Bronze (raw ingestion), Silver (cleansed & conformed), and Gold (business consumption) layers.
Tracks high-watermarks automatically in the control layer to process only new or modified records, skipping unchanged data.
Automated Type 2 Slowly Changing Dimensions (SCD2) with attribute change tracking and deterministic surrogate key generation.
Records data dependencies and structural lineage metadata into the control Lakehouse for auditability and governance.
Fabric Variable Libraries dynamically resolve environment configurations across DEV, TEST, and PRO.
Meet the Architect
Creator of the Fast Fabric Framework (F³). Dedicated to delivering robust, metadata-driven architectures that eliminate technical debt and ensure sustainable data engineering in Microsoft Fabric.
Implementation Pricing
Transparent fixed pricing for deploying the Fast Fabric Framework directly into your Microsoft Fabric tenant.
Tiers 1 & 2: Framework Implementation + Full Team Autonomy
The aim of Tier 1 and Tier 2 is to implement the framework directly in your tenant (at different levels of scale and complexity) and provide comprehensive explanations and knowledge transfer to your internal team, ensuring they are completely autonomous to maintain, operate, and evolve the architecture independently.
Tier 3: Extended Consulting & Advisory
Tier 3 adds dedicated consulting and advisory services to continue further beyond the initial framework implementation—covering custom Gold layer modeling, complex multi-tenant architectures, and strategic data engineering guidance.
Tier 1
Framework implementation for a single basic source system + complete handover for team autonomy.
- 1 Basic Source System (Salesforce, Oracle, SQL Server, PostgreSQL, MySQL, etc.)
- Up to 5 Tables (Bronze + Silver layers)
- Metadata config JSON structure & Control Lakehouse setup
- Full & Incremental loading support with watermarks
- Surrogate Keys & SCD2 dimensional handling
- Field & table lineage logging
- Operational handover & explanations for 100% internal team autonomy
Tier 2
Framework implementation for up to 3 basic source systems + complete handover for team autonomy.
- Up to 3 Basic Source Systems (Salesforce, Oracle, SQL Server, PostgreSQL, MySQL, etc.)
- Up to 25 Tables (Bronze + Silver layers)
- Metadata config JSON structure & Control Lakehouse setup
- Full & Incremental loading support with watermarks
- Surrogate Keys & SCD2 dimensional handling
- Field & table lineage logging
- Multi-environment variable library setup (DEV, TEST, PRO)
- Comprehensive architecture walkthrough & enablement for 100% team autonomy
Tier 3
Custom framework implementation plus dedicated consulting to continue further beyond implementation.
- Implementation of the F³ Framework tailored to your tenant
- Custom number of source systems and tables
- Gold layer business modeling & consumption design
- Dedicated consulting & advising services to continue further beyond implementation
- Architecture review and best-practices guidance
- Ongoing team enablement and operational handover
How F³ Compares to Traditional Ad-Hoc Projects
An honest, transparent comparison between deploying a pre-built metadata framework versus developing custom data pipelines from scratch.
Transparent Bottom Line: With F³, you do not pay for reinventing common data engineering foundations — you only pay for deployment, adaptation, and tenant integration.