Greenfield development
We build what no commercial product covers, across the full lifecycle: from the written specification to observability in production.
“[AI] empowers trained professionals to write a lot more … software that would otherwise never be written.”
Some needs no commercial product covers. A pricing engine that only exists in your business, a website no template fits (this one, for instance). That's what greenfield development has always been for.
And there's a newer reason. With the capabilities AI brings, building custom now pays off in cases where buying used to be the default.
Fair warning first, though. If a SaaS solves it, we'll say so and we won't build it.
Code is not the product
The product is the system still running years after the demo. The discipline behind that never changes, AI in the loop or not.
/**
* Four rules, none of them optional.
*
* @param stack Java without nostalgia. Spring Boot or Quarkus on Java 21. Virtual threads, records and pattern matching. Code that scales without tricks.
* @param contract The contract comes before the code. The API is agreed in writing (OpenAPI), with OAuth 2.0 and mTLS in the design, not in a patch.
* @param governance CI/CD is governance. Tests, static analysis and security checks gate every deploy. You always know what runs on your servers, and why.
* @param security Security gets compiled in. Fine-grained access control in the code, and Cloudflare's WAF in front.
* @author the same engineers, start to finish
* @since 2019
*/
public System build(Stack stack, Contract contract) { … }
Greenfield ≠ blank page
We never start from zero, and we don't let an AI do it either. For every technology frame we maintain our own working framework: project skeleton, conventions, environments, quality and deployment solved before the first line of business logic.
Pythonon Google Cloud. Services and functions on Cloud Run and Cloud Functions, with the same scaffolding for environments, credentials and tests in every project.QuarkusandCamelonDocker. Millisecond startup, integration built in, ready for any orchestrator.- Publishing to
Karaf. For platforms that live onOSGiand deploy by bundles on Karaf, without stopping the rest of the system.
You feel the difference on the second service. It reads like the first, deploys the same way, and anyone on the team can run it. Boring on purpose.
With AI, the framework comes first
Maybe greenfield is where AI pays off most. It is certainly where it derails first. On a blank page, an agent delivers in hours what used to take weeks. Without a frame, every one of those hours is fresh debt.
Our frameworks are that perimeter. The agent writes inside the rails, and an engineer signs every line that reaches main. The full story is in our AI vertical.
Your technical debt is our problem
From the first commit to operations, the same senior engineers. And if the system one day moves to your team, we see the handover through and write everything down: the knowledge stays in your house.
Let's talk, engineer to engineer.
Bring the problem as it is. You'll hear back from the person who'll write the code.
Tell us about your project