AI and agentic development
AI means two things here: what we build for our clients, and how we build.
Agents for your company
We build agents that read your documents and do the routine steps in your systems. Where your data is processed is agreed with you before any code is written, and every decision that matters stays with a person.
- Reading invoices and documents into structured data
- Assistants that answer from your own data
- Automation of routine steps inside your existing systems
We build with agents
Our own products are written by AI agents working under the supervision of our engineers. It makes us fast; what makes the result trustworthy is the discipline around it.
How the quality is kept
Rules become checks
Each rule of the code base is checked by a program before a change can be committed, not remembered by someone.
Every change is read
A separate reviewing agent reads each change against the rules before it is merged; a blocking finding stops the merge.
Evidence, not claims
Every change records the commands that were run and what they returned.
Tests for behaviour
Each behaviour is covered by tests that run before every merge.
People decide
What is built, and whether it is right, is decided by people; the agents write.