Data sources
Nightly load into the data warehouse
- Accounting06:02Done
- CRM · customers06:06Done
- Billing06:10Done
- Contact centre06:14Done
- Plan and budgetRunning

A SQL data warehouse, Power BI semantic models and reports, automation and AI agents. The whole solution – from business requirements to a report management can rely on.
NewPower BI as code with GitHub and Claude CodeWhat an end-to-end BI solution involves – from the nightly data load to the report and safe AI access. The components are live, the data illustrative.
Nightly load into the data warehouse
Before data reaches the model
Actual vs. budget, m
| Month | Actual | Budget |
|---|---|---|
| Nov | 9.5 | 9.8 |
| Dec | 10.3 | 10.1 |
| Jan | 10.1 | 10.4 |
| Feb | 10.6 | 10.2 |
| Mar | 10.4 | 10.6 |
| Apr | 11.3 | 10.9 |
| May | 11.0 | 11.2 |
| Jun | 11.1 | 11.3 |
| Jul | 11.9 | 11.4 |
| Aug | 12.0 | 11.5 |
| Sep | 12.1 | 11.6 |
| Oct | 12.6 | 11.8 |
12-month churn rate
Nothing reaches production unchecked
AI on the data warehouse, safely
I found 214 supply points. Here are the first three, personal data is masked.
| Customer | Region | Commodity |
|---|---|---|
| J*** N*** | South | Gas |
| P*** K*** | North | Electricity |
| M*** S*** | Central | Electricity |
Moving reports from .pbix to PBIP, versioning them on GitHub and developing with Claude Code – with a controlled path into production in Power BI Service.
Quality checks that don't run at the end but on every change. What fails isn't saved – and before an edit you know everything it affects.
An AI assistant that knows the structure and numbers of the data warehouse – but can't change anything, overload production or see personal data.
Interactive replicas of Power BI reports on fictional data.
A finance overview for management: where the group makes money, how it compares to budget and what exactly drove the gap.
How many customers leave, why, where – and whether retention works. A report for sales and management of an energy supplier.
Contact centre operations and call quality scored by AI – compared against supervisor scoring.

A BI analyst based in Brno. For over three years I've been building reporting at an international company – from warehouse data to the reports management decides on.
I enjoy the whole cycle: understanding the problem, designing the solution and taking it into production. Lately I build BI with AI agents – faster, with the same quality checks.
Happy to discuss BI, reporting or AI in analytics. Drop me a line.