Problem
In most of the construction projects, site layout is still a manual, time-consuming process. Workers measure, mark, and re-check lines repeatedly, which often leads to mistakes, misalignment, and wasted time. Even with modern tools like BIM, the process depends heavily on human interpretation. This lack of precision can delay schedules and increase costs. That’s exactly what I wanted to solve.
Solution
TraceBot (a robot) solves this problem by bringing digital precision directly to the construction site. It’s a robot that prints floor plans on the ground based on drawings from Revit or CAD. Using the TraceBot Revit plugin, designers can send layout data straight to the robot without manual setup. On-site, workers can control TraceBot through a simple tablet interface — to start, pause, or adjust the printing in real time. This makes the layout process faster, more accurate, and much easier, reducing errors and saving both time and effort for the construction team.
GIF of TraceBot (50cm × 40cm × 25cm)
Glymphs of Tablet Screens:
Design Approach
I followed a Triple Diamond approach, starting with research to understand real on-site challenges, then narrowing down to a clear core problem. I explored multiple ideas, selected the most promising solution, studied competitors, and finally moved into prototyping and iterative improvements. Throughout the process, I switched between diverging and converging to make sure every step stayed user-focused and purposeful.
Triple Diamond approach
Talking to Real Users
To understand the real challenges on construction sites, I interviewed eight architects who have been working in the field for over three years. I asked open-ended questions to let them share their experiences freely, and I framed my questions to uncover pain points, frustrations, and gaps in their current layout-marking process. These conversations helped me gather a wide range of insights that shaped the direction of this project.
Gathering All the User Insights
Making Sense of the Insights
After the interviews, I mapped all the insights and grouped them into themes like communication, BIM workflows, and site layout marking. When I compared these themes, the strongest and most repeated issues came from the site layout side—how plans are translated onto the floor. I dug deeper into those layout-related insights and, from there, identified one clear core problem that became the foundation for the rest of the design work.
Grouping Insights into Themes
Ideating Around the Problem
To explore solutions, I used the Crazy 8s technique, sketching eight different ideas in eight minutes to push quick, creative thinking. I then combined the strongest concepts and generated a few more refined ideas. After reviewing them against the core problem, I selected one promising direction that had the most potential to solve the issue effectively and fit real on-site needs.
Selecting a potential solution
Looking Around the Industry
To understand how others are solving this problem, I explored both direct and indirect competitors in the construction-tech space. I studied their products, captured screenshots, analyzed their design language, and reviewed how their workflows were structured. I also examined their websites and reached out to people working at some of these companies to learn more about their tools and approach. This helped me identify gaps, opportunities, and best practices that shaped my own solution direction.
Bringing the Idea to Life
To bring the solution to life, I started by sketching different ideas on paper and mapping the overall product experience flow. Once the structure felt clear, I created low-fidelity wireframes to outline the interface and interactions. After that, I explored several AI tools like Figma Make, Lovable, Cursor, and Google Stitch—to quickly generate screen variations and gather visual inspiration. I refined those AI-generated concepts, iterated on the layouts, and combined everything into a cohesive high-fidelity prototype.
Crafting the Experience
End to End User Flow
Early Prototype Testing
After creating the mid-fidelity prototype, I tested it early with five architects to validate my assumptions before investing time in polishing the design. Their feedback helped me spot usability gaps and refine the flow. I analyzed the insights, made targeted improvements, and then moved into building the high-fidelity version.
Refinement Strategy
TraceBot Revit Plugin
TraceBot Tablet
Inclusion of the golden ratio (1:1.618) in UI design of Plugin
Inclusion of the golden ratio (1:1.618) in UI design of Tablet Screens
Building the 3D Model
To design the full end-to-end experience, I also created the physical model of the robot. Every component of the design was backed by research and functional reasoning. I modeled the robot in SketchUp, refined the form through iteration, and finally 3D-printed the prototype to validate scale, proportions, and usability in a real-world context.
Future Development
Going forward, I’m exploring several enhancements for the product:
Error rectification: allowing the robot to erase incorrect prints and reprint updated layouts.
Multi-robot control: enabling one operator to manage multiple robots on a single site.
Zone-based printing: dividing large floor plans into manageable zones for faster, structured printing.
Business direction: evaluating potential business models and market opportunities for scaling the product.