For years, AI in architecture has largely focused on generating concepts and images. This workshop looks at the next step: AI agents that can work directly with Rhino and Revit through Model Context Protocol (MCP), helping architects generate geometry, interrogate BIM models, manage parameters, detect clashes, and export project data.
Across a live five-hour workflow, participants will develop a mid-rise mixed-use housing block by translating architectural constraints, including setbacks, unit mixes, cores, heights, and façade modules, into explicit rule sets that AI agents can execute and architects can audit.
🚀 You’ll learn how to:
• Set up and configure Rhino MCP and Revit MCP
• Turn spatial briefs into parametric rules
• Generate rule-checked massing and variants in Rhino
• Query and audit BIM models in Revit
• Manage family parameters, clash detection, and schedules
• Build reliable project context files for AI agents
• Benchmark GPT-6 Astra and Claude Fable 5 on identical architectural tasks
• Manage token costs, caching, and model security
The goal is not to hand design decisions over to AI. It is to establish a workflow where AI handles repetitive execution while the architect defines the rules, directs the process, and evaluates the result.
From prompt → to agent → to geometry → to BIM.
📑 Topic: GPT-Astra in Architecture: Agentic AI Modeling with Astra, Fable, Rhino and Revit
📅 Date: October 4th, 2026
🕓 Time: 14:00–19:00 GMT
🧑🏼🎓 Total Seats: 50
🛒 15% Discount for Digital Members