Cities are the most interconnected systems most of us will ever work with, and almost all of their data is open. This journey teaches you to acquire a city from OpenStreetMap, turn it into a reusable urban profile, and then model it — accessibility, site selection, livability, and price — at the resolution of individual buildings, before closing with the network analytics that run underneath all of it. Six steps, mostly vector data, one city at a time.

Most people meet geospatial the hard way, jumping right into the middle — a client request, a new library, an error message nobody can explain. This journey goes the other way: seven steps from an empty computer to geospatial-ready Python, built for people leaving point-and-click GIS and for data scientists who just realised their problems were spatial all along. Four of the seven steps are not about maps, and that is not a coincidence.

Ever since I started sharing geospatial work, the most common question has been where do I start? Now there is a proper answer: three ordered routes through the whole library — Getting Started, Urban Data Science, and GeoAI — each a sequence of courses, books, and free tutorials where every step assumes the one before it. Here is how they fit together, and which one is yours.

I've written over 13,000 prompts for geospatial Python work, and one in four was me correcting the model. Here are the five ways I actually use AI, from a 3D Manhattan built in one prompt to mining 1,281 of my own conversations, and why each one still needs a driver who knows the territory.

A while ago, shademap.app went viral: pick any place on Earth, drag a time slider, watch the shadows of every building crawl across the streets. What struck me was not just the idea — it was the speed. Every shadow tool I had met before, mostly inside desktop GIS stacks, was painfully slow: computing a single neighbourhood could take hours. And here was a website doing it in real time, in a browser.

The AI build-out has made data centers a lightning rod — for their energy appetite, their water use, and, this summer, a viral wave of before-and-after images claiming they're bulldozing forests. I ran the numbers on two of the biggest open questions, both with free, public geospatial data. Here's what the evidence actually says.

I spent years doing data science on problems that were quietly about place — without knowing GIS existed. This is the map I wish I'd had: what the field is, its sub-areas, the core Python stack, and the order to learn it in.

By now, humanity has wrapped the planet in sensors, from the smartphones in our pockets to the satellites circling the globe, generating spatial data at a previously unimaginable scale — and the capacity to make sense of it all is still catching up. GeoAI is here to help to close that gap