jeremytammik

jeremytammik / eubim

Public

AI and BIM Programming at EUBIM 2026 Valencia

18
0
100% credibility
Found May 23, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
C#
AI Summary

This repository contains Jeremy Tammik's workshop materials for EUBIM 2026, demonstrating how AI coding assistants can help create add-ins for Revit—the industry-standard building design software. The project includes practical examples that automatically generate 3D building models with walls, floors, roofs, doors, and windows. It showcases experiments with autonomous AI agents that can write, test, and fix building code in real-time, along with documentation exploring the future of AI-assisted BIM programming. The author is a recognized Autodesk technology expert who retired in 2025 after decades of teaching the Revit API community.

How It Works

1
🎤 Discover AI-powered BIM at a conference

You hear Jeremy Tammik explain how he stopped writing code himself and started using AI to build add-ins for Revit, the industry-standard building design software.

2
📁 Browse the workshop examples

You explore the repository and find ready-to-use examples showing how AI can create walls, floors, roofs, doors, and windows automatically.

3
🏠 Watch a house build itself

You see the 'Little House' command run inside Revit and create a complete 3D building with multiple floors, a roof, and openings for doors and windows—all generated by AI-written code.

4
🤖 Understand the AI workflow

You learn how AI tools like GitHub Copilot and Codex help developers describe what they want in plain language, and the AI writes the code to make it happen.

5
Choose your learning path
🌱
Start with Demo01

Begin with the simple Hello World example to understand the basics of Revit add-in development.

🚀
Explore Codex experiments

Jump into the autonomous agent loop experiments to see how AI can test and fix its own code in real-time.

6
📊 See detailed building reports

Every building the AI creates comes with a clear report listing all the floors, walls, roofs, and openings it generated.

✨ Gain confidence in AI-assisted building design

You now understand how AI can handle the repetitive coding work while you focus on the creative decisions in building information modeling.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 18 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is eubim?

This is a research playground from Jeremy Tammik (the legendary Building Coder) demonstrating AI-assisted BIM programming. It shows real experiments using GitHub Copilot and Codex to build Revit add-ins that generate 3D building elements--walls, floors, roofs, doors, and windows--without writing code manually. The repo documents his transition from human coder to AI orchestrator, including a full case study of generating an MEP HVAC system using Claude Opus 4.6.

Why is it gaining traction?

BIM programmers are watching AI coding agents mature rapidly. Tammik's work is unique because it shows the actual cost and workflow of using AI to build production BIM tools, not toy examples. The MEP HVAC case study with real usage metrics (293 API requests, cost tracking) gives concrete data on what works and what doesn't. Plus it's tied to EUBIM 2026, giving it conference credibility.

Who should use this?

Revit add-in developers curious about AI-assisted coding. BIM tool creators wanting to automate repetitive modeling tasks. AEC developers evaluating whether to integrate coding agents into their workflow. Not useful for developers needing production-ready libraries or those unfamiliar with Revit/BIM concepts.

Verdict

This is a personal research repo (18 stars, minimal structure), not a productized library. But the author's reputation and documented experiments make it valuable for anyone exploring AI-assisted BIM development. At 1.0% credibility, treat it as an interesting case study rather than a reliable resource.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.