AI is everywhere right now. Every product, every headline, and every conversation seems to involve artificial intelligence. But for those of us working in the AEC industry, I think the bigger question is: How can we actually use AI to make our day-to-day work easier?
I've spent more than 30 years working in architecture, construction, and design technology. I've seen plenty of new technologies come along, and there's always a period when everyone is trying to figure out what's genuinely useful and what's just noise.
That's where I found myself with AI. I'm not an AI expert, and I don't know how to write code. But I do know Revit. I know how templates should be built, how standards need to be applied, and how much repetitive work goes into getting everything configured correctly.
So I started wondering: Could AI help me automate some of that work?
Putting My BIM Knowledge to Work with AI
I started experimenting with Claude, a conversational AI tool, to see which parts of building a Revit template I could automate.
Instead of asking AI to simply build something for me, I explained exactly what I needed. I provided my existing template information, office standards, naming conventions, and instructions for how each tool should behave.
Claude helped me develop the Python code and file structures needed to create custom tools through pyRevit. My job was to define the requirements, test everything in Revit, identify what wasn't working, and make adjustments.
It wasn't a one-and-done process. There was plenty of back and forth, and the first result was rarely right.
But what started as a single automation button eventually grew into a toolkit of more than 160 tools, including capabilities for configuring project settings, creating system families, generating views, and organizing sheets.
I didn't suddenly become a programmer. I used my BIM experience to guide AI through the parts I didn't know how to do myself.
Where Else Could AI Help Your Team?
You don't need to develop an entire toolkit to start finding value in AI. There are plenty of smaller tasks worth exploring.
For example, AI can help you:
-
Check BIM standards: Review view and sheet naming conventions or identify inconsistencies in shared parameter files.
-
Support quality control: Organize Revit warnings or identify missing room information.
-
Improve documentation: Turn project requirements and notes into structured BIM execution plans or reference guides.
-
Develop automation: Get help creating AutoLISP routines for AutoCAD, Python scripts for Dynamo, or structured files for Navisworks coordination.
These are all examples of work where you already understand what needs to happen. AI can help with the technical steps, but you still need to provide the direction.
AI Still Needs You to Know What You're Doing
One of the most important things I've learned is that AI can sound completely confident and still be wrong.
While developing my Revit tools, Claude told me that dimension types depended on text types and needed to be created in a particular order.
I knew that wasn't correct because I understand how Revit works. I had to explain the mistake and correct its assumptions.
That's why I always emphasize testing the results. AI doesn't know your firm's standards, your project history, or why certain decisions were made. And it doesn't always have accurate information about the software you're using.
You also need to follow your firm's AI policies and be careful about what project or client information you share with these tools.
Start Small and Build From There
If you're wondering where to begin, my advice is simple: Pick one repetitive task that you understand well and that you can easily check.
Explain what you want to accomplish, give AI the rules it needs to follow, and see what it produces. Then test it, correct what doesn't work, and try again.
You might not save time on your first attempt. I certainly spent plenty of time developing and refining my tools. But once you establish a process that works, you can build on it.
AI hasn't replaced my BIM knowledge. It's helped me use that knowledge in ways I couldn't before. And I think that's where the real opportunity is for our industry.
See How I Put AI to Work in Revit
In my on-demand webcast, I walk through the tools I developed, demonstrate how they work in Revit, and share more examples of how AI can support everyday AEC workflows.
Jim Stoneberger, AEC Solutions Consultant, IMAGINiT Technologies
