Reference
From goals to working solutions
Technology is not a result in itself. We help our customers identify a problem worth solving, build a durable solution to it and make the change part of how the organisation works.
Our references show the whole chain, from strategy and definition to data, AI and technical implementation, and on to deployment and operational change.
Case study · Western Uusimaa Wellbeing Services County
From fragmented data production to a unified data platform
What did they want to achieve?
A reliable and scalable way of producing information, so that each new report does not create its own definition, its own query and its own truth.
How was it done?
Data production began to be developed with software engineering methods. The work moved from SQL to PySpark and Python, long data pipelines were broken down into testable transformation functions, and automated testing, CI/CD and data contracts were introduced.
What changed?
Releases are more controlled and can be made more often. At the same time, a documented and repeatable way of working emerged that does not depend on any single person or team.
Moving to the Data as Software model has fundamentally changed the way we approach data production. We want to understand data and produce knowledge.
Other recent work
Below are a few examples of problems we have solved together with our customers.
From data architecture to an actionable roadmap
- Goal
- To turn a fragmented current state into a shared target state and a manageable path forward.
- Delivery
- The data architecture and its phased roadmap were built to take into account the organisation's capabilities and operational change, not only the technology.
- Impact
- Technical and organisational changes could be brought into the same plan and carried out in controlled phases.
Research data in a common data model
- Goal
- To make health data from different source systems usable in a consistent and repeatable way for research and analytics.
- Delivery
- Data was transformed into the common OMOP data model, the data platform and data production were developed, and the maintenance and documentation of vocabulary mappings were built into the process.
- Impact
- Data can be produced more consistently and repeatably for secondary use.
A data platform for a regulated environment
- Goal
- To build a data platform in an environment where information security, traceability and controlled production use are baseline requirements.
- Delivery
- We designed the technical architecture and developed the platform's tools and key processes, from access control to monitoring, releases and recovery.
- Impact
- The data platform can be used and developed in a controlled way, even in a demanding and regulated operating environment.
From AI readiness to use cases that can be delivered
- Goal
- To understand where AI is worth using and what production use requires.
- Delivery
- We assessed use cases, data, ownership, the technical environment and the solution's whole lifecycle together with operational needs.
- Impact
- AI development can be focused on use cases with a real need and realistic prospects of reaching production.
From situational awareness to operational management
- Goal
- To bring scattered resource and operational data together to support operational management and preparedness.
- Delivery
- We defined how data sources are combined and harmonised, and how the whole of data, operating models and responsibilities is built.
- Impact
- The situational picture can become part of management rather than just a technical view.
Do you have a problem worth solving?
We do not start from technology. We start from what you want to achieve, and look for the right way to use data, AI and technology to get there.