With preliminary product vision and research, Dwellci AI is struggling with a functional MVP to validate concepts and test out user feedback.

The Solution

As the only designer on the team, I built key flows and AI features to support project and study creation, as well as AI model generation and optimization. The MVP received 4.88/5 in preliminary user testing.

Dwellci AI

An AI-led Transformation of “How Architects Might Work?”

Architects are known to go through complex development stages, where they gather client and site requirements, brainstorm multiple ideas, explore feasibility, and hop onto different softwares like AutoCAD, Revit, Rhino to create polished models and rendering.

AI is coming in to change how they work. Starting with Dwellci AI, aiming to empower the early conception phase with AI — making the process faster and freer.

The Problem

How might we guide users to input data for AI generated models?

The Challenge

AI generated models depend on data input, such as site information, building programs, as well as potential constraints like construction codes. In a study creation process, I needed to figure out a way for this process to go smoothly and intuitively.

Gathering AI generation model requirements

Through analyzing the workflow of a building construction, competitive analysis, and consultation with the founder, I realized there are 4 essential steps before a model can be generated by AI.

Comparing user-AI interaction flows

With the information about model building, I explored multiple flows of inputting information and generating models with AI. Their main differences lies in how our AI model can understand user input and transfer that input into precise models.

Flow 1 — Only input programs -> AI generates model -> User adds constraints

Flow 2 — Input constraints + metrics -> AI generates model ✅ (Dev efficient & user effective)

Flow 3 — User directly describes models -> AI generates model -> User edits

Guiding Users through the Process

Based on the flow, the user data input flow was divided into 3 key stages, with more details at each stage.

Feature—Scalable Site Selection

Users can select their building site for once, and repetitively build studies on the same site.

Feature—Guided User Input and AI Models

Data input are divided into multiple categories for easy input and simplified onboarding.

Current User Problems with Complex Workflows

Jumping back and forth between different softwares cause a few issues in architects’ development process: it takes extra time importing and exporting files in different softwares, sometimes having to recreate details caused by incompatibility. Besides, keeping track of different files and manually updating one view due the change in another view adds to the extra work.

Switching Smoothly between Views & Metrics

As a result, I optimized the mid-fi wireframe into the features of a universal view switch bar and metrics reference panel.

Architecture software usage guide: archimash.com/articles/architecture-software-guide/

How might we integrate multiple workflows into one solution?

The Challenge

Currently, architects juggle with multiple software to process different types of files, including 2D floor plans, data files, 3D models, and client requirements. To streamline their workflow, the challenge here is to reduce steps and provides integrated solutions.

Exploring Best Integration Structure

Feature—Real-time Updated Views & Metrics

Users can switch between different views originally achieved on multiple softwares, and enjoy real-time synchronized data between different model changes.

Feature—Chat with AI to Modify Models

Users can chat with AI to describe what changes they want to make to the model, and select between different versions of changes to finalize a concept.