Geospatial Data Science Essentials: Quick Guide to Your First GeoAI Agent

Build a working GeoAI agent - from scratch, on your laptop, in an afternoon.

This free Quick Guide walks you through building a conversational geospatial AI agent: ask it questions about Manhattan in plain English ("which neighbourhood has the most bars?", "how many cafes are within 500 metres of the Empire State Building?") and it answers by calling real spatial tools in a reasoning loop — no black box, no framework magic.

You'll wire up five spatial tools, a language model, and a hand-built ReAct loop, and watch the agent decide which tool to call at each step. Everything runs on free, open data (OpenStreetMap) and a free LLM API — no GPU, no satellite data, no cloud bills.

What you'll build:

• Five spatial tools (geocoding, radius search, neighbourhood ranking, and more)
• A spatial index for fast queries over 24,000+ points
• A from-scratch agent loop you fully understand end to end

Written for anyone who knows a little Python and wants to see how a language model wires to real spatial data — assembled piece by piece, not imported as a black box.

This guide is one chapter of Geospatial Data Science Essentials: 101 Steps to GeoAI from Scratch — the complete build-everything-yourself GeoAI curriculum.