AI Engineer · Netherlands
The model proposes. The code decides. A human signs off.
I build AI systems that ship into regulated businesses.
About
I'm the engineer at a five-person AI consultancy in the Netherlands, which in practice means I build the thing, deploy it, write the documentation, sit in the client meeting, and then explain it to the board.
In four months I've put an assistant live inside a client's own Azure tenant twenty days after the first commit, rebuilt an insurer's risk-assessment process as a tool where the AI proposes and a human decides, and built our entire back office - invoicing, e-signing, bank reconciliation, VAT returns - because buying five subscriptions was the worse option.
Before that: an MSc in Artificial Intelligence at the University of Amsterdam, and three published papers on text simplification and fairness in graph neural networks.
I build with agentic coding tools day to day, and I say so to clients rather than letting them find out later.
Skills
- AI / ML
- Large language models · agent loops with tool use · RAG · provenance-tracked extraction · prompt design · LLM-as-judge evaluation · PyTorch · Transformers · Azure AI Foundry · OpenAI and Anthropic APIs
- Backend
- Python · FastAPI · SQLAlchemy · Node.js · Postgres · SQL · REST APIs · golden-file and unit testing
- Frontend
- React 19 · TypeScript · vanilla JS · hand-built SVG data visualisation
- Cloud / DevOps
- Microsoft Azure (App Service, Entra, Bot Service, OpenAI) · Docker · GitHub Actions · Railway · Cloudflare · OAuth and secret handling
- Microsoft 365
- Graph API · Teams bots · Power Automate · Office Scripts
- Compliance
- GDPR and EU AI Act assessment · EU data residency · retention rules · DPA and sub-processor review
- Languages
- Dutch · English · Japanese