You’ve seen the rise of AI tools in architecture—colleagues sharing concept images, animations, and client-ready visuals generated in minutes. But where do you start? If you’ve felt left behind or overwhelmed by the pace of change, this course is your entry point into the future of architectural design.
In this future course, you’ll go from sketch to finished visual assets—learning how to generate, control, and refine images, diagrams, and even videos using ComfyUI and open-source generative tools. You’ll get step-by-step guidance to install and run workflows locally or online, even if you don’t have a powerful GPU. Discover how to sketch with AI, fine-tune materials, lighting, and structure, and adapt outputs to your daily tools like Revit, SketchUp, ArchiCAD, or 3DS Max. Learn how to upscale your visuals, produce animations for free, and build a practical AI workflow that enhances—not replaces—your creativity.
This course isn’t about hype. It’s about giving architects the confidence and tools to work smarter, stay relevant, and lead the transformation of our field.
Week 1 – Installation, Foundations & First AI Creations
Objective: Understand the ecosystem, set up tools, and generate your first visual output.
AI for Creatives: The Landscape
What AI can/can’t do for designers
Tool comparison: MidJourney, ChatGPT, Stable Diffusion, etc.
Choosing the right tool by use case (text, image, video, code)
Setup & Access
Local (ComfyUI for GPU) vs. Online (Runpod, Vast.ai, Hugging Face, etc.)
Finding and using free models/workflows
First Prompt & Image
Write basic prompts
Generate your first AI sketch with Stable Diffusion
Week 2 – Design with Precision: Control All Elements of Your Image
Objective: Learn to guide AI outputs and match specific visual goals.
Control AI Outputs
Use ControlNet with sketches, depth maps, edges
Create or export guides (SketchUp, hand-drawings)
Refining Composition
Adjust style, lighting, material, perspective, and scale
Explore texture tiling and spatial logic
Style Matching & Iterations
Reference image tools (IPAdapter, Chatgpt, Flux Kontext)
Control the season, lighting, texture, architectural style, …
Week 3 – Integration, Presentation & Animation
Objective: Merge AI outputs with real workflows, polish visuals, and explore video animation.
Workflow Integration
Overlay AI visuals on CAD/SU/3DS/Archicad views
Export diagrams, create moodboards and exploded views
Post-Production & Presentation
Upscale with Real-ESRGAN, Photopea basics
Match branding/drawing styles, prep for screen and print
Intro to Video & Final Polish
Tools overview: RunwayML, Kling, CapCut, etc.
Storyboarding, transitions, and exporting polished video clips
Tho-Phuoc NGU is a former federal architect turned AI consultant, currently working as Product Owner of Arcana Academy. After 7 years serving at the Ministry of Foreign Affairs leading international embassy projects, he shifted toward technology, blending his diverse skills in design, technologies, and generative AI.
As self-taught developer and strong communicator, he also built two Educational YouTube channels (Music & Generative AI). Building the academy was a natural step in his mission to make innovation more accessible through practical, real-world learning.
Join weekly live sessions with feedback and expert mentorship. Pay Once.
Choose a starting date
9 Remaining seats
Missing a Class ?
No worries! All live courses will be recorded and made available on-demand to all students.
9 Remaining seats
Join weekly live sessions with feedback and expert mentorship. Pay Once.
Choose a starting date
9 Remaining seats
Missing a Class ?
No worries! All live courses will be recorded and made available on-demand to all students.