AI for Architecture and Design Capstone Exercise — AI for Architecture and Design

What you'll learn

Apply skills from across this course — concept visualization, pitch narrative, space planning, material research, compliance research framing, documentation, client communication, competitive research, and IP disclosure — to a single realistic project scenario
Produce a coherent, client-ready package of design deliverables using AI tools at each stage, with your own professional verification built in at every checkpoint
Self-assess your package against the professional-liability and originality standards established throughout this course

This course has covered AI-assisted concept generation and visualization, client presentations and pitch narratives, space planning support, material and sustainability research, code and zoning compliance research, project documentation and specifications, client communication and revision management, competitive and portfolio research, and IP and originality considerations. The capstone asks you to coordinate all of it into a single client-ready package under a realistic timeline and a realistic set of constraints.

The thread running through every lesson in this course has been the same: AI accelerates the drafting, structuring, and research layer of design practice, and your professional judgment and verification remain non-negotiable at every checkpoint where a client, a code official, or the public will encounter the result. This exercise rewards a package that shows that judgment explicitly, not just AI-generated polish.

Capstone Exercise

Boutique Wellness Studio Build-Out: Client-Ready Package in Two Weeks

Context

You are the project lead at a small interior design and architecture practice. A new client is opening a 3,200-square-foot boutique wellness studio (yoga, physical therapy, and a small retail area) in a ground-floor commercial unit of a converted early-20th-century warehouse building. The client wants a warm, natural-materials aesthetic that respects the building's industrial character, has a firm opening-date target that leaves you two weeks to deliver a client-ready proposal package, and has not yet confirmed the site's zoning allows the physical therapy use as a matter of right in this district.

Your Task

Using AI tools of your choice (naming which tool you used for each piece), produce a client-ready proposal package with the following components. Component 1: a five-element concept image prompt (typology, mood, materials, lighting, viewpoint) grounded in the actual brief, plus a one-paragraph description of the resulting concept direction. Component 2: a four-part pitch narrative (problem, vision, solution, proof) for presenting this concept to the client. Component 3: a structured adjacency matrix for the program (yoga studio, PT treatment rooms, reception/retail, staff support) with at least one explicit separation requirement. Component 4: a short material comparison table for the flooring system, including at least one sustainability claim you have explicitly verified against a primary source, with the source noted. Component 5: a two-to-three sentence zoning and use-classification research note for the PT use, written so that it correctly frames the finding as a starting point requiring verification with the Authority Having Jurisdiction, not a compliance determination. Component 6: a short client-facing disclosure statement (two to three sentences) explaining, in plain language, that AI tools were used in concept development and how the firm verifies AI-assisted work before it reaches the client.

Your notes (optional)

Deliverable

A six-component client-ready package (concept prompt and direction, pitch narrative, adjacency matrix, verified material comparison, correctly-framed zoning research note, and client disclosure statement), with the AI tool used for each component named, and with your own verification step explicit wherever this course has identified verification as required — material claims, zoning framing, and AI tool disclosure in particular.

Quick check

The capstone says the zoning research note is not judged on whether it reaches the right answer about the physical therapy use. What is it judged on instead?

Select one answer.

Key takeaways

A professional AI-assisted design workflow moves through the same checkpoints regardless of project type: structured AI input at each stage, followed by your own verification wherever a claim, a compliance question, or an attribution could create risk if wrong.
The quality difference between a professional AI-assisted deliverable and a generic AI output is almost always the specificity of the input and the presence of an explicit verification step, not the sophistication of the AI tool used.
Compliance research, material sustainability claims, and project attributions are the three areas in this course where an unverified AI output creates the most serious downstream risk, and each requires the same discipline: treat the AI output as a starting hypothesis, not a finished answer.
Disclosing AI tool use to clients, proactively and in plain language, is a trust-building practice, not a liability admission — most clients respond well to a firm that is transparent about its process and confident about its verification standards.