Using AI Assistants to Generate ArcGIS Pro Python Scripts for Automation
Eric Pimpler
For years, the barrier to automating GIS workflows has been the same: you had to learn Python, then learn ArcPy’s extensive library, then debug cryptic error messages—all before you could automate a single task. Many GIS professionals never made it past the first hurdle.
AI assistants like ChatGPT, Claude, and Gemini have fundamentally changed this equation. Now you can describe what you want in plain English, get working code, and iterate quickly when something doesn’t work. Your expertise in GIS workflows becomes the valuable skill—you know what needs to happen, and the AI handles the syntax.
But here’s the reality that the hype often misses: AI-generated code doesn’t always work on the first try. The real skill isn’t just prompting—it’s knowing how to iterate between code generation and execution until you get working results.
In this article, I’ll walk through a realistic workflow using a file geodatabase with feature classes typical of municipal GIS operations. You’ll see exactly how the conversation between you and an AI assistant unfolds, including the errors that occur and how to recover from them.
For more information about this topic and ArcGIS Pro please consider our ArcGIS Pro Bootcamp coming up in Denver later this February.