AI engineering for Amsterdam, from Science Park to Zuidas.
RAG, agents, evals, observability, streaming UIs. We ship AI products from Groningen, on the same business day as the Amsterdam AI cluster. Spin-outs from UvA Science Park, post-Series-A scale-ups bolting on AI features, internal-tools teams replacing back-office work: we've shipped the shapes.
What we build
RAG over your own data
Document ingestion, chunking, hybrid retrieval, reranking, citation traces. pgvector first, vector DBs second, evals from day one.
Agents you can ship to real users
Tool calls, structured outputs, retries with budgets, deterministic test paths. Agents that run unattended with audit trails.
Streaming + cancellable UIs
Token-by-token rendering, cancellable streams, optimistic state. The UX visitors now expect.
Costs + latency observability
Token tracking, model A/B, latency dashboards. AI is a production system; we instrument it that way.
Where this fits
You're an Amsterdam scale-up bolting an AI feature onto an existing web app and the team has shipped CRUD, not LLMs.
You're a Science Park spin-out with a strong model and a weak product; we close the gap.
You're a B2B SaaS adding an AI assistant on internal data and you want it observable before it ships.
Tech stack
- TypeScript
- OpenAI / Anthropic
- RAG pipelines
- pgvector
- Vercel AI SDK
- BullMQ
Want this for your team?
30 minutes to scope what you need. No pitch deck, no obligation. We tell you straight whether Stacklane fits.
Book a Free Call