Finland · enterprise data & technical delivery
I design how data moves from source systems into usable products, and build reusable workflows to implement and verify those designs.
My professional work is in financial and insurance data platforms. Through independent projects, I also build and test tools for AI-assisted engineering, with explicit work state, execution permissions and recovery.
Senior Solution Analyst at OP Pohjola Chapter co-lead
Make source behaviour and consumer needs explicit in models, quality rules and data-product lifecycles.
See the professional contributionProfessional experience
My work connects business requirements, source-system behaviour and the teams building and using data products. At OP Pohjola, the focus is insurance data and cloud modernization. Earlier consulting work at Teradata covered wealth-management data and regulated reporting.
OP PohjolaSenior Solution Analyst · 2022–present
TeradataBusiness analytics consulting · 2018–2022
Information & data-product architecture
Designed information and data-product models connecting source systems, legacy warehouses and cloud platforms with engineering, reporting and source-system teams. Defined semantics, temporal behaviour, keys and quality expectations to guide implementation.
Integration & lifecycle design
Defined data-product lifecycle patterns covering quality, access, service readiness and retirement. Shaped migration approaches around reporting and analytics consumers, and evaluated Databricks integration options across CDC, federation, snapshots and Delta-based consumption.
AI-assisted data engineering
Developed and introduced an AI-assisted documentation and lineage workspace that consolidated source mappings into traceable specifications and modelling inputs. Demonstrated the approach across the organisation, where other teams used it as a reference.
Executable validation
Designed executable SQL regression and validation checks with source lineage and repeatable baselines, used by the delivery team to assess target data. Earlier work combined requirements, semantic design and final quality assurance for regulated customer reporting.
Chapter co-lead
Co-lead solution-analysis competence development through hands-on workshops, shared requirements practices and AI-assisted engineering examples. Peer learning and reusable material help colleagues apply the methods in their own work.
Independent engineering
These are personal projects with public implementations. They show how I approach integration, validation and recovery. Each links to a concrete example and states the scope of its evidence.
Source descriptions checked .
Work coordination & acceptance
What survives when the agent changes?
I built a coordination and settlement layer that preserves work identity, evidence and acceptance when an agent stops or another worker takes over.
The handover proof covers a bounded workflow. Whole-estate capture coverage and broader deployment qualification remain open.
Source data to reporting
Can one model support several useful decisions?
I built a platform that turns heterogeneous household sources into shared models, reporting and scenarios. The finance workflow follows a statement from ingestion through reporting to a saved scenario.
The synthetic workflow demonstrates the implementation. It does not establish current household outcomes or savings.
Model & agent execution
Which work may run, where, and who may read the result?
I built a configurable model and agent platform that separates model choice, execution, data placement and result access. Jobs retain their configuration and authority context.
Synthetic exercises have a defined scope. Real installations require qualification against their actual models, configuration and environment. The repository-session worker launch is not released.
I also operate a Kubernetes homelab and keep public notes on migrations, recovery and architectural decisions.
How I work
I prefer systems where responsibilities are explicit, important decisions survive the meeting, and a failed change has a planned route back.
Understand source behaviour, consumer needs and ownership. Establish which system records each important fact.
Use small implementations and executable checks to compare alternatives and expose assumptions before they spread.
Record what was requested, changed, accepted and rejected. Keep permissions, evidence and operating state understandable.
Make setup, validation, handover and recovery repeatable, so the next person can work without reconstructing the whole project.
Where I want to contribute
I am interested in work with room to investigate the problem, make design decisions and stay involved through implementation and use.
Enterprise teams
Responsibility for information models, integration choices and data-product lifecycles, working with business and delivery teams from the initial problem through operating use.
Data & AI platform companies
Field and forward-deployed work that combines technical discovery, prototyping and implementation: understand the customer's constraints, test a viable approach, and leave a reusable solution with clear limits.
Outside the technical work
I live in Finland. Away from data and infrastructure, I spend time on ancient history, cooking, travel, making things, 3D printing, and game ideas that usually acquire an economy before they acquire a title.
Read about the workshop