Dinofni

Microsoft Fabric · Azure Synapse

Data platforms on Microsoft Fabric and Azure Synapse, built to run in production.

Lakehouses, warehouses and pipelines on Microsoft's data stack, designed and built from the first ingestion job to the Power BI model your business reads every morning.

Discuss your project See services
OneLakeMedallion lakehouse
  1. SourcesSQL Server · SAP · REST APIs · files
  2. Bronzeraw Delta tables, append-only
  3. Silvercleaned, conformed, deduplicated
  4. Goldstar schema for reporting
  5. Power BIDirect Lake semantic model
Incremental refresh · change data capture
  • Fabric Lakehouse
  • Fabric Warehouse
  • Data Factory pipelines
  • Dataflows Gen2
  • PySpark notebooks
  • Delta Lake
  • Azure Synapse Analytics
  • T-SQL
  • Power BI / Direct Lake
  • Azure DevOps
  • GitHub

Services

Development on the Microsoft data platform

Fabric

Lakehouse and warehouse builds

OneLake lakehouses and Fabric warehouses laid out in bronze, silver and gold layers, with Delta tables partitioned and optimized for the way your reports query them.

Synapse → Fabric

Synapse to Fabric migration

Dedicated SQL pools, Spark notebooks and Synapse pipelines moved into Fabric with a tested cut-over plan, parallel runs and reconciled numbers before the old workspace is switched off.

Data Factory

Ingestion pipelines

Data Factory pipelines and Dataflows Gen2 that load from SQL Server, SAP, REST APIs and files, using incremental loads and change data capture instead of nightly full reloads.

Spark · T-SQL

Spark and SQL development

PySpark notebooks and T-SQL transformations with tested logic, sensible Spark pool and capacity settings, and query tuning when jobs get slow or expensive.

Power BI

Semantic models

Direct Lake and import models on top of the gold layer, with clean star schemas, row-level security and measures your analysts can reuse.

Governance

Platform setup and DevOps

Workspaces, capacities and access control set up properly, Git integration, deployment pipelines across dev, test and prod, and capacity monitoring to keep costs predictable.

Approach

How an engagement runs

  1. 01

    Assess

    Review of your sources, current Synapse or SQL estate, data volumes and reporting needs, with a capacity and cost estimate.

  2. 02

    Design

    Target architecture, workspace and security model, naming and layering conventions, agreed before any build starts.

  3. 03

    Build

    Pipelines, notebooks and models delivered in short iterations from your Git repository, each one demonstrated on real data.

  4. 04

    Hand over

    Runbooks, documentation and working sessions with your team, so the platform runs without outside help.

About

Focused on the Microsoft data platform

Data Platform Engineering works only on Microsoft Fabric, Azure Synapse Analytics and the Azure services around them. That focus means faster delivery and fewer surprises, because the edge cases of capacities, Spark pools, Delta tables and Direct Lake are already known.

Every engagement ends with code in your own Git repository, deployment pipelines your team can run, and documentation written for the people who will maintain the platform after handover.

Every engagement delivers

  • All code in your Git repositoryPipelines, notebooks, SQL and semantic models under version control.
  • Deployment pipelinesPromotion from dev to test to production without manual copying.
  • Architecture and runbooksWritten for the people who operate the platform day to day.
  • Knowledge transferWorking sessions with your engineers before handover.

Contact

Start a project

Describe your current data setup and what you want to achieve. You'll get a reply by email.