Who this is for

You work with data — spreadsheets, dashboards, maybe some Python — and you keep meeting problems that are quietly about place. Or you are new to code altogether and want the shortest honest route into the field. Either way, this journey assumes nothing.

Assumes: nothing — no Python, no GIS, no maths beyond a spreadsheet.

By the end you can

• Write clean Python for real data work: pandas, NumPy, matplotlib, and the habit of inspecting before you trust.

• Explain what makes data spatial — coordinate systems, vector and raster, Tobler's law — before you touch a library.

• Load, reproject, join, and map geospatial data with GeoPandas, Shapely, and OSMnx.

• Read a map, a spatial join, or a model output critically, and know the three ways every spatial dataset is a little bit wrong.

How to follow this journey

• Go in order. Each step assumes the one before it — the route is the prerequisite chain, not a ranking.

• Steps 1 to 3 are free: the field map, a 13-minute video, and the complete Data Science Essentials PDF. Do those before you buy anything.

• On paid steps, Learn more opens the course or book page; Start the course / Get the book goes straight to Gumroad, where most items have a free sample.

• Going deeper is optional and unordered — a shelf, not a route. Come back to it after step 7, or whenever a step raises a question.

The route

7 steps · 3 free · roughly 4 hours of video plus two books

1. The Map of Geospatial Data Science: How to Actually Learn It

The field, drawn as a map. Where data science, geospatial analytics, and GIS overlap, why most people walk past the middle of it, and where you actually start.

Read →   Free article

2. One Prompt, 46,000 Buildings

The destination, before the first line of code: a full 3D Manhattan built in one live session with an AI assistant — and what the assistant still needed a driver for.

Watch →  ·  Read the write-up* →   Free video · 13 min

3. Data Science Essentials: 101 Practical Steps in Python

The ground floor. Install Python, repair a deliberately broken dataset, draw every chart twice — the honest version beside the misleading one — and learn when to skip the model. The complete 246 pages are free; the bundle adds the notebooks and data.

Learn more →  ·  Free PDF →  ·  Get the full bundle →   Book · free PDF

4. Python Foundations for Geospatial Data Science

The same foundation in video, with two things the book leaves out: the workflow habits that make an analysis reproducible instead of lucky, and a whole chapter on writing prompts that get correct code first time.

Learn more →  ·  Start the course →   Course · ~3 h

5. I Mined My Entire Claude History

Chapter 10 of the course, applied for real: a data-science pipeline over 1,281 of my own AI conversations, and the rule that came out of it — Python counts, the model judges.

Watch →  ·  Read the write-up* →   Free video

6. How to Think Spatially

No code, deliberately. One hour on what makes data spatial, why nearby things break normal statistics, how a globe gets flattened onto a screen, and the traps that catch professionals. Every concept points to the exact step in the next book where you build it.

Learn more →  ·  Start the course →   Course · ~1 h

7. Geospatial Data Science Essentials: 101 Practical Python Tips and Tricks

The capstone, and the foundation for everything else on this site. 101 techniques from a single point to spatial regression, on the current stack — GeoPandas 1.x, Shapely 2, OSMnx 2, H3 4. Finish this and every other journey is open to you.

Learn more →  ·  Get the book →   Book

Going deeper

Optional, any order.

Connecting the Dots

Networks, data, algorithms, and how the systems around you decide what you see next — the ideas before the code, with no code at all.

Learn more →  ·  Get the book →   Book

The map is wrong and nothing threw an error

How spatial analysis fails quietly, and the tool that catches it before review does.

Watch →  ·  Get the skill →  ·  Read the write-up* →   Free video + free Claude skill

The World Map with Many Faces

Every flat world map lies somewhere; here is where, drawn in Python.

Read* →   Tutorial*

Map Projections in Cartopy

The one projection tool the core stack does not cover.

Read* →   Tutorial*

Geospatial File Formats

Shapefile, GeoPackage, GeoJSON, GeoParquet — what each is good at and where it breaks.

Read* →   Reference*

* Part of my premium Substack plan.