Client:
Cellverse, iHuman Research
Industry:
Computer Vision, B2B
Role:
Founding Product Designer
Timeline:
Sep 2024 - Jul 2025
Human-in-the-Loop Labeling, Job Orchestration & Quality Control
Reshaping the way structural biologists work with scientific data
AI-Assisted Data Ops for Imaging
Overview
End-to-End
| Web UIUX Design
| Data Visualization
| Agile Development
At a computer-vision AI startup, we built an intuitive web platform that brings cutting-edge cryo-EM algorithms to structural biologists. I designed an AI-assisted data-ops interface includes user managemant, labeling, quality scoring, and re-run—that turns messy images into reviewable, traceable, re-runnable tasks, cutting time to results and improving model-ready data. Adopted by 200+ researchers, it now powers 2M+ processing runs.
Current Challenges
Traditionally, Cryo-EM/Cryo-ET image processing and analysis
for structural biologists looked like this :
Cluttered with Windows
Complex Parameter Logic
Opaque Processing Status
Time-Heavy Debugging
Behind on AI Adoption
Collaboration Unfriendly
current software platform: imod
Current Challenges
Researchers' goal is to analyze Cryo-ET images with tools like IMOD, but cluttered windows, confusing parameters, and uncertain processing create frustration, while failures demand slow debugging and outdated software offers neither AI integration nor modern collaboration.
Technical Context
Meanwhile,
new AI algorithms for Cryo-EM pipelines are ready to be integrated :
DRACO —— Cryo-EM Foundation Models for Automated Pipeline (developed by team Cellverse)
How Might We
Adopt emerging technologies to design a new Cryo-EM data processing platform with greater usability and collaboration?
Outcome Impact
Adopted by 200+ active users with highly positive feedback
Since launched on Dec 31, 2024
200+ active users
2M+ tasks processed
Who Did I Work With
Collaborative End-to-End Delivery in an Interdisciplinary Team
I collaborated with the Cellverse team who developed the DRACO algorithm, facing the challenge of entering an entirely unfamiliar field. Through proactive learning and close collaboration with colleagues from diverse backgrounds, I not only delivered usable design specifications that received strong user feedback, but also applied design thinking and skills to educate and give back to the team.
This experience greatly strengthened my confidence in thriving within interdisciplinary teams.
What Key Steps I Took?
Knowledge Transfer — PRDs
Defining how the new interface should look and function:
clustered stakeholders’ descriptive needs into clear design to-dos
From the seminar, I gathered client-user needs (structural biology researchers) and the development team’s technologies to be integrated, and clustered them into design modules.
User Flow Redesign
From scattered points to create a new workflow narrative
Suggested a new user flow that addresses key user needs by enhancing task efficiency and collaborative management, integrating login, user management, a dashboard task center, and dual Cryo-EM/ET pipeline processing.
What’s Different Now?
1.Centralized task management and visualization
Probelm:
The old UI was cluttered with modals and frequent window switching, leading to poor efficiency.
Solution:
Introduce a centralized dashboard and navigator-guided modular features to enhance overall system integration and streamline user workflows.
create task
view task
2.Streamlined parameter settings
Probelm:
Broken parameter logic and lack of annotations created high cognitive load for novice users.
Solution:
Redesigned as a linear form submission flow with user-based reclassification, default values, and parameter explanations, simplifying setup to under 4 steps with easy editing and navigation.
3.Partial Re-run & Failure Isolation
Probelm:
Users see only overall task progress, with no visibility into subtasks. Failures at the subtask level degrade results, but the interface offers no way to locate issues or re-run in time.
Solution:
Built subtask progress visualization with responsive layouts, added quality score visualization and partial re-run support to accelerate error recovery.
4.Enabled easy user management and collaboration
Probelm:
The outdated software lacks support for modern collaborative lab workflows.
Solution:
Introduced registration, login, and role-based access control, and project sharing mechanisms.
5.Make AI outputs visible and adjustable
Interactive module visualizing AI results with threshold control and side-by-side comparison.
Final Layout
An One-stop User-friendly Platform for Fully-Automated Cryo-ET Data Pre-processing and Visualization
Efficient task management that adapts to user habits:
preserved frequently used key parameters, while deprioritizing low-value image details into a collapsible list to reduce cognitive load.
Seamingless AI integration:
In the noise reduction workflow, simplifying front-end parameter tuning into a single bar control. This enhanced user sense of control while enabling direct comparison with the original image.
User Management
Easily bring lab partners into the loop and collaborate.
Status
Enable on-time human intervention.
Options
Simplify traditional complex editing.
How I Grow From Here?
Collaborative End-to-End Delivery in an Interdisciplinary Team
I collaborated with the Cellverse team who developed the DRACO algorithm, facing the challenge of entering an entirely unfamiliar field. Through proactive learning and close collaboration with colleagues from diverse backgrounds, I not only delivered usable design specifications that received strong user feedback, but also applied design thinking and skills to educate and give back to the team.
This experience greatly strengthened my confidence in thriving within interdisciplinary teams.
Product Innovation @Siemens Healthineers
B2B | Strategy Design | Product Design | Case Study
User Acceptance Testing
@Danone
B2C | UX Research | User Interview | Insights Alignment
AI Product Innovation
@UC Berkeley
Entrepreneurship | VUI | UXD | Emerging AI Tech | Marketing
UXR & Inno-Design
@Siemens Healthineers
Field Visit | Contextual Inquiry | Insights Alignment| Concept Design