About
A BI developer who owns the whole data path.
I'm a Microsoft Certified Power BI Data Analyst with 3+ years building the reporting businesses run on — and the warehouses underneath that make it trustworthy.
THE SHORT VERSION
My core discipline is Data Analytics & BI: Power BI semantic models, DAX, SQL reporting and the dashboards leadership actually opens. That is where I started, and it is still most of what I do.
What makes that reporting hold up is everything around it. I build the warehouse underneath — medallion layers in Microsoft Fabric, Azure Data Factory pipelines, a governed star schema — so a dashboard is never limited by data quality somebody else owns. And I add the predictive layer on top: forecasting, churn and risk scoring, and the AI automation that acts on the result.
I've delivered 38+ projects across GitHub, LinkedIn and Upwork. The standard is the same every time: understand the data before you build anything on top of it. That is why the reporting reconciles — and why anything built on it can be trusted.
HOW THE WORK FITS TOGETHER
Reporting is the job.
The rest is what protects it.
Most BI developers depend on a pipeline someone else owns. Most AI builders have never had to make a number reconcile. Owning the path from source system to dashboard — and beyond it — is what stops reporting breaking for reasons outside your control.
The reporting layer
Where most of my work sits: semantic models, DAX, SQL reporting, row-level security and the dashboards decisions get made from.
- Power BI reports · dashboards
- Power BI Service workspaces · apps · refresh
- Semantic modelling star schema · DAX
- SQL & T-SQL reporting queries
- Row-level security one number, one definition
What holds it up, and builds on it
The warehouse that keeps reporting trustworthy, and the forecasting and AI automation that extend it once the data can carry them.
- Microsoft Fabric medallion warehouse
- Azure Data Factory ETL pipelines
- Machine learning forecasting · scoring
- LangChain & n8n AI automation
PRINCIPLES
How I work
Diagnose before building
The first question is never which tool. It's what decision is being held back, and what's actually in the data that's stopping it.
Own the whole stack
No handoffs between specialists. I understand your data from the source system to the dashboard to the agent acting on it, so nothing gets lost in translation.
Ship production, not demos
Everything I build is meant to run unattended, be maintained, and survive real usage — validated at every layer rather than impressive once.
Want to work together?
Tell me about the data problem you're facing. I'll tell you honestly whether I'm the right fit.