Madrigal.co
Contact us

Legacy-to-lakehouse migrations

Migrate legacy systems into cost-effective lakehouse architectures.

Modernize your data foundation with future-ready architectures that scale beyond today, reduce complexity, optimize infrastructure costs, and support analytics, inference, and growth.

Connected data foundation and contextOperational sources enter a layered foundation, a context network connects its entities, and resulting decisions return to operations.

Tools we use

Lakehouse engineering for migration, pipelines, governance, analytics, and model-ready data.

Unified data workloads across OneLake, Power BI, Data Factory, and enterprise analytics environments.

Success Stories

Nova Southeastern University logo

Nova Southeastern University

  • Modernized university reporting with a Microsoft Fabric lakehouse.
  • Migrated 133 Power BI reports and rebuilt supporting data workflows.
  • Created a scalable foundation for growing data and analytics needs.
Hudson Advisors logo

Hudson Advisors

  • Migrated 33 stored procedures to Databricks using Spark SQL and PySpark.
  • Rebuilt data pipelines with Delta tables and automated orchestration.
  • Achieved a reported 30% cost reduction compared with the legacy environment.
Puerto Rico Government logo

Puerto Rico Government

  • Connected legacy SQL Server data to Microsoft Dataverse.
  • Enabled a Power Apps CRM for managing business incentive processes.
  • Streamlined data integration and workflows using Azure Data Factory and Power Automate.

Data services built for migration, scale, and inference

From source systems to a governed lakehouseRead records from operational systems. Capture changes and check data quality. Model and govern the data in a lakehouse.Operational sourcesCapture+ validateLakehouseModelled + governedSources, validation and lakehouseRead records from operational systems. Capture changes and check data quality. Model and govern the data in a lakehouse.Operational sourcesCapture and validateGoverned lakehouse

Data Engineering Solutions

Migrate legacy data systems into modern, cost-effective lakehouse architectures designed to scale beyond today.

Explore Data Engineering
A request routed to a model, with an evaluated responseReceive a request with the context it needs. Select a model to process the request. Return the response and evaluate its quality.Request+ trusted contextModel BModel AModel CModel routingResponseResponse evaluationModel routing and response evaluationReceive a request with the context it needs. Select a model to process the request. Return the response and evaluate its quality.Request withtrusted contextModelroutingModel BModel AModel CModel responseEvaluation

Inference Models

Turn structured data into predictions, recommendations, and operational decisions.

Explore Inference Models
Business facts linked through entities and rulesBring together business records and documented knowledge. Relate entities and apply business rules. Retrieve context with its supporting facts and rules.Facts + knowledgebelongs togoverned byAccountPolicyOrderContextOrders related to accounts and policiesBring together business records and documented knowledge. Relate entities and apply business rules. Retrieve context with its supporting facts and rules.Business recordsand knowledgeOrderbelongs togoverned byAccountPolicyContext for a query

Context Layers / Ontology

Connect systems, entities, and business logic into a shared context layer your teams and models can use.

Explore Ontology

Modern data systems should not break every time the business grows.

Growth should not force teams to rebuild pipelines, duplicate logic, or keep paying for infrastructure that cannot scale with the business.

From isolated reporting to a shared lakehouseBefore, ERP, CRM, and files each require separate extraction, business rules, and reports. After, those same sources share ingestion and a lakehouse: Bronze preserves raw data, Silver validates it, and Gold prepares it for business use. Governance spans all three layers. Shared data supports analytics, context, and inference.

Before

Each system needs its own extraction, business rules, and reports.

ERP
Manual extraction
Local business rules
ERP report
CRM
Manual extraction
Local business rules
CRM report
Files
Manual extraction
Local business rules
Files report

After

Source data is refined in shared layers and reused across workloads.

ERP
CRM
Files
Shared ingestionBatch & change data

Lakehouse

  1. BronzeRaw

    Preserve source records

  2. SilverValidated

    Clean and standardize

  3. GoldBusiness-ready

    Model for business use

GovernanceAccess, lineage & policy across every layer

  • Analytics
  • Context
  • Inference

Meet the team

Mateo Madrigal

Mateo Madrigal

Technical Lead

Data Engineer with 7+ years of experience delivering scalable, cloud-based data solutions across multiple industries. Focused on modern data architectures, pipeline optimization, and cost-effective data platforms.

LinkedIn
Jake Henao

Jake Henao

Sales Manager

Sales professional with 7+ years of experience connecting business needs with tailored data solutions. Focused on client relationships, strategic planning, operational efficiency, and delivering measurable business value.

…and

Data Engineers,Machine Learning Engineersand Data Analysts

working with us across companies.

Contact Us