Solo: design, data pipeline, backend, map UI
Fahsai
Fires, wind and PM2.5 on one map of Thailand's burning season, with an AI explanation for any station on any day. Designed and built solo, with Claude Code as my pair programmer.
Product engineer · frontend & design at the core
I build data- and map-heavy products, from idea to production.
Eighteen years of turning complex data into interfaces people can read and use. These days with AI-native workflows that cover ground that used to take a small team.
Fires, wind and PM2.5 over mainland Southeast Asia on 5 April 2026, rendered by Fahsai.
Data: NASA FIRMS, Open-Meteo (CC BY 4.0). © Mapbox © OpenStreetMap © Maxar
- Core
- Frontend architecture, maps, data visualisation, interaction design
- Also
- APIs, Postgres, ingestion jobs, deployment, AI features
- Growing
- Infra: CI, containers, cloud
- Based
- Utrecht (CET), remote and async
- Open to
- Freelance, contract or employed
Selected work
Sole frontend architect, geospatial & data visualisation
Restor.eco
The Google Maps of nature, for the world's largest network of restoration sites. The first version was slow and buggy, so a small team rebuilt it from scratch. I was the sole architect of the geospatial and data visualisation frontend.
How I work with AI
In the product
Code decides, the model explains
Fahsai's “What's causing this?” traces the air back 72 hours and classifies the day before Gemini writes a word. Sixteen hand-checked station-days and an eval script keep it honest. The same backend is an MCP server, so other models can use it too.
In how I build
A harness, not prompts
Claude Code writes a lot of my code, inside guardrails: my conventions as skills it loads before writing or reviewing, hooks that gate on lint and tests, and pull requests kept small enough to actually review.
In a business
AI adoption at a policy consultancy
At Aster I set up the org's Claude workspace: shared context per organisation, project and consultant, Agent Skills for work like Dutch legal citation checks and monthly reporting, and a remote MCP server for hours data with its own OAuth flow. Next up: a monthly policy monitor.
From the lab
All experiments →Got a product where the data is the hard part?
I'm looking for my next product engineering role on a data- or map-heavy product. Based in Utrecht, working remote.
dev@joostkiens.com