Fusion Twin is a web platform for nuclear fusion simulations.
2024-2025
Founding Product Designer
I joined as the sole Product Designer and worked closely with the development team.
Researchers worked across disconnected files, parameters, and specialist tools, making simulation setup and data exploration difficult to follow.
As the sole Product Designer, I turned complex scientific requirements into clear workflows, interface patterns, and a product structure that worked for both researchers and students.
I designed a unified workspace for managing files, configuring simulations, tracking status, and exploring scientific results.
Testing showed around 30% better navigation efficiency across selected tasks, while the browser-based workflow made the product easier to access without local installation.
• Gold Winner — Vega Digital Awards 2024 (Science & Technology).
• Bronze Winner — Summit International Awards 2025.
• Strategic Funding Catalyst: Designed a professional-grade interface that helped secure key R&D funding by demonstrating commercial scalability in a $10B+ market.
Researchers often had to move between different tools to prepare data, configure simulations, and explore results. The problem was not only that the interface was complex. The whole workflow was fragmented.
The main question became:
How can we bring these steps into one clear experience?
We had two important user groups. Experienced researchers needed detailed controls, reliable data, and flexibility.
Students and less experienced users needed more guidance and a safer way to experiment.
So the product had to support experts without overwhelming beginners.
.png)
I used interviews and workflow analysis to understand how people worked with simulations.
too much setup before useful work could begin
files and simulations lived in different places;
comparing results took too many steps;
collaboration was harder when tools were disconnected.
The opportunity was not to simplify the science.
The next step was deciding where everything should live. I grouped features around real researcher tasks and simplified the paths between data, simulation setup, and analysis. This created a clearer information architecture for the growing product.
There were many possible improvements, but we could not solve everything at once. I prioritized the areas that had the biggest impact on the core workflow:
managing scientific files
configuring simulations
check simulation states
understanding results
I prioritized the areas with the biggest impact on the core workflow: file management, simulation setup, status tracking, and results exploration.
Before building detailed wireframes, I sketched a few early ideas to explore layout, hierarchy, and the main actions without spending time on visual details.
Before moving into high-fidelity design, I explored the most complex workflows with wireframes.
At this stage, I focused on:
hierarchy
navigation
order of actions
information density
This helped solve structural problems before adding visual detail.
.png)
Early wireframes helped validate hierarchy, parameter grouping, and system feedback before moving into high fidelity.
The dashboard became the main entry point into the platform.
From one place, researchers could access files, simulations, saved outputs, and visualizations.
The goal was to make the current state of their work easy to understand at a glance.
.png)
Fusion research often works with complex datasets, including HDF5 files.
Researchers needed one reliable place to upload, organize, inspect, and reuse data across simulations.
I designed a file workspace that made data part of the product workflow instead of something users had to manage outside the platform.
.png)
Creating a simulation can involve many parameters and technical decisions. Experts needed flexibility, but showing every option at once would make the flow harder for less experienced users. So I designed a guided setup that kept important controls available while making required steps easier to follow.
.png)
Researchers often work with more than one simulation at the same time.
They needed to quickly understand what was: draft, queued, running, completed, or failed.
I designed clear states and a simulation list so users could track progress without opening every simulation individually.
.png)
Scientific values are rarely useful in isolation. Researchers needed to understand how files, parameters, and outputs were connected. I designed a mapping experience that made these relationships visible and easier to configure before analysis.
.png)
Running a simulation was not the final goal. Researchers needed to compare experimental data, simulation results, and several runs. I designed configurable chart layouts that allowed users to explore multiple datasets in one workspace and quickly see where results differed.
.png)
In many consumer products, simplifying means hiding information. That would not work here.
Researchers needed access to detailed scientific data. So instead of removing complexity, I focused on organizing it.
Important information appeared first, while deeper controls remained available when users needed them.
Fusion Twin was still growing, so I did not want every new scientific workflow to require a completely new interface.
I created reusable patterns for: tables; forms; file states; simulation statuses; graphs; spacing; responsive behavior.
This gave the product and development team a more consistent foundation for future features.
.png)
Fusion Twin brought file management, simulation setup, parameter mapping, and results exploration into one browser-based workflow.
Testing showed around 30% better navigation efficiency across selected tasks.
The platform also created a reusable foundation for future scientific workflows and helped make the product easier to access without complex local installation.
Fusion Twin taught me that a complex product does not always need to become simple.
The more useful question is:
Which complexity belongs to the science — and which complexity is created by the tool?
.png)