About
Samuel Dias
I'm a data professional who started on the ground floor — literally learning SQL and Python on my own time — and worked my way into a data team by building something good enough to get noticed. That instinct hasn't changed: I still care more about whether a number can be trusted than about how a dashboard looks.
What I Do
I build the layer between raw data and business decisions — from Databricks/PySpark pipelines to the executive dashboards people actually use to make calls. My strongest work tends to start with a question nobody had quite asked yet: not "what's the number," but "can we trust the number, and why does it keep changing."
How I Work
I default to ownership end-to-end — ingestion, modeling, governance, and the final report — rather than handing off pieces along the way. I use a PR-based Git workflow even for BI work, because I think analytics deserves the same rigor as software. And I'd rather flag a metric I don't trust yet than ship a clean-looking number I can't defend.
What I'm Learning
Right now: dbt and Airflow, to formalize the orchestration and testing discipline I've been applying informally for years — plus enough French to widen where I can work. Long-term, I'm moving deliberately from BI toward Analytics Engineering and Data Engineering, without losing the business fluency that BI forced me to build first.