Inspiration We wanted to make 3D design more accessible by combining human creativity with AI. Our vision was to let users transform ideas into 3D designs through simple text, voice, and sketches without being limited by traditional CAD complexity.

What it does Draw3D is a human-AI collaborative CAD platform that enables users to create and refine 3D designs through text, voice, and sketches. It combines Three.js, FreeCAD, and Blender with AI assistance, desktop interaction, and optional WebXR support.

How we built it We built Draw3D using JavaScript, Vite, Three.js, and Node.js. It integrates FreeCAD and Blender for native modeling, OpenAI as the primary AI provider, and Vertex AI as backup. Google Cloud Run, Cloud Storage, and Secret Manager power its cloud infrastructure.

Challenges we ran into Our biggest challenges were integrating browser-based 3D rendering with native CAD tools, supporting multimodal user input, coordinating AI services, and building secure, isolated cloud workspaces.

Accomplishments that we're proud of We built a working browser-accessible Draw3D platform that brings together AI assistance, interactive 3D design, native CAD tools, cloud infrastructure, and multiple input methods in one environment.

What we learned We learned that effective AI-powered design requires more than automation. It needs iterative feedback, human control, seamless tool integration, and an architecture that connects AI with real 3D modeling workflows.

What's next for Draw3D We plan to improve AI-to-CAD automation, deepen FreeCAD and Blender integration, enhance multimodal understanding, introduce real-time collaboration, and expand immersive WebXR experiences.

Built With

Share this project:

Updates

Submission history