About — the long version

Hi, I'm GiGi.

Domain architect, data enthusiast, and reluctant pragmatist about Copilot. Based in the Netherlands.
Gigi Levens at her home-office desk in a mustard sweater, a notebook of pipeline sketches open in front of her and a BI dashboard on the screen behind.
At the desk · pipeline sketches 2026 · The Netherlands
— Origin story

I came into data through the side door — communication.

I grew up in Curaçao and trained as a communications specialist, with a research master's spent analysing how people actually talk — reading sentiment at scale, coding thousands of messages by hand, learning that a number is a message before it's a fact.

Web, SEO and marketing analytics pulled me closer to the data, until one day I was building the pipelines instead of just reading what came out of them. The instinct never changed: dig into the messy problem, find the smallest lever, push gently, and watch the system rearrange itself.

Today I work on the Microsoft data stack — Fabric, Power BI, Purview, Azure Data Services — with the people who own the data inside larger and semi-public organisations. I design architectures, audit ones already in flight, and write essays about the parts nobody documents.

One thing that shapes everything here: I care how the writing reads. Every essay is drafted and edited by hand, sentence by sentence, until it sounds like me — dry asides, em dashes and all. That punctuation isn't an AI tell; it's a habit from years of working across languages, and I'm not giving it up. If the prose comes out a little particular, that's on purpose.

What I stand for.

Three commitments I've put on every project for the last decade. They sound obvious. The reason I keep writing about them is that almost no one actually does the work.

01

The bridge between tech and the business is the work.

Senior BI/data specialists earn their fee in translation, not implementation. I gather requirements, turn them into clear technical plans, coordinate Agile or waterfall delivery — and I'm never afraid to roll up my sleeves and write the SQL alongside the team.

02

End-to-end means end-to-end.

Years of Power BI work, including the unglamorous part: training stakeholders to actually act on what the dashboard says. The better part of a decade on the Azure data stack, from on-prem migrations to greenfield cloud builds. I see the whole pipeline because I've built every segment of it.

03

Architectural rigour is a kindness to your future self.

Five years shaping enterprise data architectures, mostly to DAMA-DMBOK standards. The point isn't compliance — it's that governable, scalable, maintainable systems are the only ones still useful three years later.

The professional journey, abridged.

The full version is on LinkedIn. The interesting parts are below — the moments where something I thought I knew turned out to be wrong.

2026
Domain Architect · Douane, via Data Talents
A step up in altitude — from solution architecture to a whole domain.

Started in August 2026, after a spring sabbatical — the first proper gap in thirteen years. The shift: from designing one system well to shaping how an entire domain's architecture holds together.

2024
Solution Architect · ProRail, via Data Talents
Architecting for a national rail programme — and translating AI risk for boards.

Safety-critical infrastructure by day; advisory work on governance and the EU AI Act alongside it. The lesson that keeps repeating: the hardest part is never the technology, it's the agreement underneath it.

2022
Data Architect · retail & governance
Cut a BI estate's infrastructure cost by 44% — by deleting, not buying.

Power BI at scale, a broken warehouse rescued without a rebuild, and the first hands-on Fabric work once it landed in 2023. Earned the DAMA data-management credential somewhere in the middle.

2020
Lead Data Platform · Scrum Master
Led a platform team through the modern-data-stack hype — and a pandemic.

Synapse, medallion, the data-mesh debate; Product Owner on a European Commission project under GDPR. Learned that you build the team before you build the platform.

2018
Azure IoT Engineer · smart buildings
Made a building talk — sensors, streams, and a digital twin.

Stream Analytics, Event Hub, Databricks, Kusto, and guest lectures on teaching data to city-makers. The edge taught me how much data lies to you before you clean it.

2017
ETL Developer · enterprise finance
Crossed from marketing analytics into real data pipelines.

BIML, SSIS, PowerShell, Data Vault, and an on-prem warehouse hauled to Azure. Discovered that documenting a migration so humans can follow it is half the job.

2013
Communications & marketing analytics
Started reading data as a message, not a number.

Web, SEO and Google Analytics for small clients; a research master's spent coding thousands of messages in Python and SPSS. Curaçao to Amsterdam, with a comms degree and a growing suspicion that the data was where the real story lived.

Certified, for the curious.

— A handful of the badges; the full set lives on LinkedIn.
A
Azure Data Engineer Associate
Microsoft · DP-203 · Current
D
Power BI Data Analyst Associate
Microsoft · PL-300 · Current
F
Fabric Analytics Engineer Associate
Microsoft · DP-600 · Current
S
Professional Scrum Master I & II
Scrum.org · Current
P
Professional Scrum Product Owner I
Scrum.org · Current
D
DAMA-DMBOK Practitioner
DAMA Intl. · 2022
B
BDS Data Science · Pro & Consulting
Big Data Science · 2021
P
Purview Data Governance Specialty
Microsoft · Beta

The stack, today.

— What I reach for first, what I reach for cautiously, and what I keep on the shelf.
— Platform · daily
  • Microsoft Fabric
  • Power BI
  • Azure Data Lake (Gen 2)
  • OneLake
— Governance
  • Microsoft Purview
  • ● Glossaries & lineage
  • ● Stewardship workflows
  • ● DAMA-DMBOK 2 framework
— Modelling & query
  • Semantic models & DAX
  • ● Tabular Editor 3
  • ● Notebooks (Python · Spark)
  • ● Synapse remnants, where alive
— Cautiously
  • Copilot in Fabric/Power BI
  • ● Data Activator
  • ● Real-Time Hub
  • ● Metrics Layer (Public Preview)

Producing a podcast, programming an event, or want to compare notes on data, AI, and governance?

LinkedIn/in/gigilevens
Based inThe Netherlands
CurrentlyDomain architect · Douane, via Data Talents
Connect on LinkedIn