Join me while I build GeoScribble, a draw the country from memory app!

This is a story about how I build apps. It's about a process I've been refining that enables quality and speed. I want to walk you through exactly what I did because it's repeatable, and I think more people can and should be building prototypes before meetings instead of just talking about ideas.

Large Language Models don’t actually "remember" your conversations. Every message you send includes the entire conversation so far. When that gets too long, applications have to make choices about what to keep, and sometimes that goes sideways.

This is what happens when you let an engineer use AI to try to play Jingle Bells. Spoiler: it's bad. But also kind of wonderful?

New year, new tool! Our first episode of 2026 tells the story of how I built the Accessible Map Agent, a new way to make maps understandable for everyone.

My dentist sent me a Wrapped this year. Apparently, I was in the top 15% of flossers in their practice...

I've been talking to my computer. Like, out loud. Full conversations. I use the dictation feature on my MacBook to tell Claude what I want instead of typing it, and I can't believe it took me this long to figure this out.

You can send satellite imagery to an LLM, ask it to "draw a red box around every swimming pool," and then extract those red boxes as vector features. No GPU required. No domain-specific training. Just an API call and some image processing.

I try to explain the different AI tools in ArcGIS that can classify or extract information from imagery.