I build and maintain the data layer that sits between your source systems and the people who need numbers: warehouses, pipelines, models and dashboards. I have been doing this since 2014, first in-house, now as a freelancer. My default is the smallest setup that does the job well.
What I do
Data warehouse setup and modelling. Design and implementation of a warehouse on Azure Synapse, BigQuery or Databricks, modelled with dbt. Includes naming conventions, testing, documentation and a deployment process your team can run without me.
Pipelines and orchestration. Ingestion and transformation jobs with Airflow, Azure Data Factory or plain Python in containers. Monitoring and alerting so failures are seen before the Monday meeting.
Dashboards and reporting. Power BI and Metabase dashboards built on a modelled layer, not on raw tables. Data marts, refresh automation, theming and the handover so analysts can extend them.
Analysis. Descriptive, exploratory and predictive analysis in Python or R when a question needs an answer before it needs a dashboard.
Audit and cleanup. A review of an existing stack: what breaks, what is slow, what nobody understands any more, and a prioritised plan to fix it.
How an engagement works
Most projects start with a short discovery call, followed by a written scope with a fixed price or a day-rate estimate. For ongoing work I offer a monthly retainer with a set number of days.
Once we start, you get:
- a Calendly link to book calls at a time that suits you;
- a Kanban board (Trello or your own tool) where progress, blockers and deadlines are visible at all times;
- a shared folder (Google Drive or your own) for documentation and presentations;
- code in your repository, documented, from day one.
I keep calls on Tuesdays where possible so the rest of the week is uninterrupted work. I work remotely from Berlin and travel to clients in Germany when it helps.
Tools I work with
dbt, Azure Synapse, Azure Data Factory, BigQuery, Google Cloud Composer, Airflow, Databricks, Docker, Python, R, SQL, Power BI, DAX, Metabase.
I am tool-agnostic. If your stack is different, the approach is the same.
Get in touch
Send a short note through the contact form or on LinkedIn with what you are trying to achieve and roughly when. I reply within two working days.