Independent Data Architect · Belgium

Data platforms built to last, not just to launch.

I design and build data architecture on Microsoft Fabric — from lakehouse foundations to the reports your business actually trusts. And when you are ready for AI, I help you take the first steps without the hype.

Start a conversation See what I do
Microsoft Fabric specialist Databricks & Power BI AI enablement
Data Sfear

What I do

Three ways I tend to be useful

Short engagements or the long haul — as architect, as builder, or as the person who helps your team make the call with confidence.

01

Data architecture

A platform design that fits your organisation: scalable, governed, and understandable by the people who have to live with it.

  • Target architecture & roadmap
  • Lakehouse and medallion design
  • Data modelling & governance
  • Migration from legacy platforms
02

Build & delivery

Hands-on implementation. Pipelines, models and reports that reach production and keep running after I leave.

  • Fabric & Databricks engineering
  • Semantic models in Power BI
  • Orchestration & CI/CD
  • Cost and performance tuning
03

AI first steps

A pragmatic entry into AI tooling: where it pays off, where it does not, and how to get there safely.

  • Use-case discovery & triage
  • Copilot & assistant rollout
  • Data readiness for AI
  • Guardrails and adoption

Main focus

Microsoft Fabric, end to end

Fabric moves fast and covers a lot of ground. I keep up with it so you do not have to — and I know which parts are ready for your production workload today.

Microsoft Fabric Databricks Power BI Azure SQL Python
OneLake & lakehouse foundationsOne copy of the data, shortcuts instead of duplication, and a layout that still makes sense next year.
Pipelines, notebooks & warehouseIngestion and transformation with the right tool per job — not everything hammered into one.
Direct Lake semantic modelsPower BI reports that stay fast at scale, with metrics defined once and shared everywhere.
Governance, security & capacityWorkspace structure, access, sensitivity, and a capacity bill you can explain to your CFO.
Deployment pipelines & CI/CDGit-backed workspaces and repeatable releases, so a change stops being an event.

Also available for

Your first steps with AI tooling

Most organisations do not need a moonshot. They need one or two useful things running well, with people who understand what is happening under the hood.

Step 01

Find the real use cases

A short scan of your processes to separate what AI genuinely improves from what merely sounds impressive in a demo.

Step 02

Get your data ready

AI is only as good as what it can reach. Usually that is an architecture question before it is a model question.

Step 03

Roll out with guardrails

Tooling, access and clear rules of engagement — plus the training that makes your team actually use it.

About

One architect. No layers in between.

I am Frederik Aerts, the person behind Data Sfear. I work as an independent data architect, which means the person doing the thinking and the person doing the building are in the same conversation.

My work sits where architecture meets delivery: designing a platform that holds up, then staying close enough to the implementation to make sure it actually does. I care about clarity — a model your analysts understand beats a clever one they avoid.

Based in Belgium, working with organisations that want their data platform to be an asset rather than a maintenance project.

IndependentDirect contact, no account manager in between.
Architecture & hands-onDesign it, build it, hand it over properly.
Fabric-firstDeep focus on one platform, fluent in the neighbours.
Flexible engagementsAdvisory days, project work, or a second opinion.

Contact

Let us talk about your data platform

Tell me where you are and where you would like to be. A first call costs nothing and usually clears things up quickly.

info@datasfear.be LinkedIn