Insurance platforms, SaaS architecture, applied AI and fintech data work, built and shipped for clients in the Netherlands.
Every engagement below went to production and stayed there. Insurance applications that hundreds of thousands of people move through, a SaaS platform used daily by thousands, and platforms we designed, built and handed over end to end. The pages behind each one set out what the client needed, what we built and what it changed.
The scale these systems run at, and what is different in them because our team was there.
People move through the insurance application flow our team builds, releases and owns
More employers that flow now supports, opening it to clients it could not serve before
Reporting use cases organisations now serve themselves, taken off the development backlog for good
Years at Qfact, from 2018 to 2024, on one microservices platform thousands of people use every day
The engagements on this page were delivered by our principal engineer, Marten den Heijer, working inside client teams. Several of them predate the founding of Starlapse B.V. in 2024, and the experience belongs to the person rather than to the company.
Every technology listed here was used to deliver something on this page. Pick one to see only the projects built with it.
A high volume insurance journey, a SaaS platform we led from 2018 to 2024, an AI system reading real user documents, and a fintech data warehouse built at startup pace.
people move through the flow that carries a.s.r.'s incoming insurance applications
We rebuilt and hardened the aanvraagstraat, the main entry point for every incoming insurance application at one of the largest insurers in the Netherlands. It serves hundreds of thousands of users.
Read the a.s.r. application flow case studyat Qfact, from 2018 to 2024, on one microservices SaaS platform used every day by thousands of people
A microservices platform for complex analysis workflows, built and led from 2018 to 2024. We introduced measurable sprint velocity where there was none and removed whole classes of recurring bugs.
Read the Qfact SaaS platform case studyapplied in production, not in a slide deck: large language models reading real user documents
A system that reads user uploaded documents, extracts what matters and generates project suggestions from it, built on large language models.
Read the QIntelligence content analysis case studydata warehouse tables modelling fiscal year reporting, replacing manual analysis with automation
Integrations with Exact Online and e-Boekhouden, plus the data warehouse that turns what they return into automated annual reports.
Read the Boekjaar data warehouse case studya.s.r. is one of the largest insurance companies in the Netherlands. Our team has worked inside its scrum team since 2024 on the platform that sells and services its policies.
We rebuilt and hardened the aanvraagstraat, the main entry point for every incoming insurance application at one of the largest insurers in the Netherlands. It serves hundreds of thousands of users.
We designed the release process for the platform that carries every incoming insurance application at one of the largest insurers in the Netherlands, ran it, and chaired the ceremonies around it. Releases became routine.
A full platform move from Sitecore and Vite to headless Contentstack and Nuxt.js, delivered while the website stayed open to visitors.
We restructured a Vue component layer that had grown past the point of clean change, put automated tests around it, and rolled the result across an entire insurance platform without interrupting live campaigns.
More coverages and more employers supported in the tool customers use to work out what their insurance costs, used with the application flow by hundreds of thousands of people.
Six years permanent, from 2018 to 2024, and a further freelance engagement on a B2B SaaS platform used daily by thousands, covering architecture, analytics, event sourcing and applied AI.
A microservices platform for complex analysis workflows, built and led from 2018 to 2024. We introduced measurable sprint velocity where there was none and removed whole classes of recurring bugs.
A system that reads user uploaded documents, extracts what matters and generates project suggestions from it, built on large language models.
The indexing and full text search foundation underneath QIntelligence, built on Python and Elasticsearch and designed to outlive the requirements it started with.
A performance focused redesign of analysis functionality, with configurable reporting that took 25 or more frequently requested use cases off the development backlog entirely.
Event sourcing for business logic persistence, giving exact visibility into what changed, when and why. It met government compliance requirements.
A service that watches for unintended behaviour, finds the data it corrupted and repairs it, before a user has to notice and report it.
Shorter engagements where the whole system was ours: the data model, the integrations, the APIs, the interface on top and the deployment underneath.
Integrations with Exact Online and e-Boekhouden, plus the data warehouse that turns what they return into automated annual reports.
A centralised platform for a residential care provider: backend, frontend, database design, APIs, authentication and interface, all built from nothing.
Tell us what you are building and where it is stuck. A short introduction call is enough for us to say whether we are the right partner for it.
Book a 15 minute introduction