About Me

Designing an AI-Powered NASA Rover Interface for Lunar Travel.

Streamlining Insurance Compliance for Mental Health Practices.

Designing an AI-Powered NASA Rover Interface for Astronauts.

At a Glance

As lead designer for a NASA SUITS team, I created an interface for a lunar rover to help astronauts during high-pressure missions. I turned overwhelming streams of data into a clear, prioritized display that and an intelligent navigation and task system. To keep the pilot’s hands free, I architected in an AI assistant that simplifies data monitoring and task flows which we got to test at NASA facilities with real astronauts.

My Role

Lead UX design, defining the design system and user flows. Helped with feature prioritization, defining interaction mechanisms and user testing strategy.

4 months

2 UX Designers

1 Project Manager

8 Developers

2 UX Researchers

My Role:

Product Designer

Founding Designer

1 Product Designer (Me!)

1 Product Manager

1 Data Scientist

3 Developers

2 Business Strategists

Results

Result

Result:

We got a

6.6/21

NASA-TLX score based on 82 user tests (lower=better)

We got a

6.6/21

NASA-TLX score based on 82 user tests (lower=better)

1 of 11 student teams chosen to test at NASA

12 Hour decrease in monthly clinical documentation time

We got a 6.6/21 NASA TLX score based on 82 user tests

We got a 6.6/21 NASA TLX score based on 82 user tests (lower = better)

Designing an AI-Powered NASA Rover Interface for Astronauts.

OUR BRIEF

Design a System for High-Stakes Lunar Missions.

The Problem:

Our challenge was to manage a massive stream of data while accounting for the unique constraints of a lunar mission. Our goal was to turn a passive cockpit into a smart partner that tracks both the rover’s health and the external crew’s safety in real-time to minimize driver cognitive load.

Solution Preview→

Final Design Preview →

RESEARCH

Exploring the data astronauts need to stay safe and focused during lunar missions.

Research Strategy →

Challenge
We couldn't get input from our target users until test week at the end of the project.

Solution
I relied on secondary research, design patterns and NASA mentor input to guide my designs.

Understanding Data Types →

Through conversations with NASA mentors and research into relevant data types, I organized the chaos of raw telemetry into a clear hierarchy which I used for my designs.

Claims page makes users review claim form to reduce errors

Exploring Similar Systems →

Investigating high-density interfaces helped me understand design patterns for balancing intricate telemetry with the interface clarity required for lunar missions.

How might we ensure life-critical vitals always command attention without distracting from the mission at hand?

Our Direction →

How might we ensure life-critical vitals always command attention without distracting from the mission at hand?

Our Direction →

How might we…
ensure life-critical vitals always command attention without distracting from the mission at hand?

How might we…
simplify the search for a lost vehicle so that navigating treacherous lunar terrain feels intuitive?

PLANNING AND ITERATION

Translating NASA Specifications into a Robust and Adaptive System.

Deciding System Form →

Option 1: One central screen

Option 2: UI projected on windshield

Option 3: Two screens; nav and data

Screen is within reach.

More room for physical controls.

Limited space for the UI.

Easy alert access.

Touch interaction is challenging.

View of terrain obstructed.

Plenty of room for data/ features.

Data/features divided by function.

Ergonomic strain for second screen.

figjam link

Defining Operational Logic & AI Integration →

Mapping the crew's complex procedures highlighted key roles for the AI Assistant and integration between the astronaut’s suit and the rover's interface.

Wire-frames to high-fi's →

See Iteration

KEY DESIGN DECISIONS

Limiting cognitive load with color scheme and AI design.

#29292C

Background

Card background

#3A3A41

Danger

Warning

Safe

Low-glare

For grouping info

Fits mental models

Designing for glare and alert visibility→

I designed for dark mode to limit the increased solar glare on the moon and reserved saturated colors for coding data states to ensure alert visibility.

*Prototyped in Cursor

Designing a context-aware AI assistant→

I considered the main use cases of the AI assistant, implementing suggested common actions and accounting for error states.

USER TESTING

Testing the system on 89 student users and NASA employees.

Testing on 89 users →

Testing Results & Procedure

NASA engineer feedback.

Experiencing the real rover.

Testing at the Johnson Space Center.

Stepping into the user's shoes.

NASA designer feedback.

Astronaut feedback.

Adapting to feedback →

Made hazard and POI Plotting more visible

Made the tracker 2d to avoid confusion

Removed the confusing controller feedback

Updated search signal flow to better communicate autonomous search

Moved speed, throttle and roll/pitch closer to windshield

Made hazards circular instead of polygonal for easier plotting.

Removed x's from medium alerts

Adjusted the warning boundaries to reduce warning fatigue

Added an AI assistant to the data screen

Added a driver supplementary screen

Added a lidar detection modal to help with driver spatial awareness

This modal would be placed right below the windshield

SOLUTION

A comprehensive navigation and data system for busy drivers.

*This mockup was partially generated by AI

AI Assisted Navigation and Task Management →

The final AI-assistant guides users through complex tasks and fluidly handles actions like point-of-interest navigation. This flow robustly accommodates every user need with strong visual feedback.

1

Intelligent alerts, data and trends →

The final system surfaces critical alerts on the main navigation screen to save users' cognitive resources for the mission. For deeper analysis, users can instantly switch to a comprehensive data screen featuring advanced trends, visualizations, and metrics.

2

SOLUTION

A comprehensive navigation and data system for busy drivers.

SOLUTION

A comprehensive navigation and data system for busy drivers.

SOLUTION

A comprehensive navigation and data system for busy drivers.

The Results →

Achieved a 6.6/21 NASA-TLX score in initial user testing, with further improvements expected from recent iterations.

Received positive feedback from reviewers and astronauts, who noted excellent usability and zero safety-critical issues.

SOLUTION

A comprehensive navigation and data system for busy drivers.

Here is me presenting Practility at Iltacon 2025

We even got our logo on "stress brains"!

FINAL THOUGHTS

Key Takeaways

The most complex UX challenges often only surface when interacting with a fully implemented system.
High-stakes systems often require breaking standard design patterns to prioritize safety and accuracy.
Partnering with researchers early ensures testing targets the most critical, ambiguous user flows.

The most complex UX challenges often only surface when interacting with a fully implemented system.
High-stakes systems often require breaking standard design patterns to prioritize safety and accuracy.
Partnering with researchers early ensures testing targets the most critical, ambiguous user flows.

The most complex UX challenges often only surface when interacting with a fully implemented system.
High-stakes systems often require breaking standard design patterns to prioritize safety and accuracy.
Partnering with researchers early ensures testing targets the most critical, ambiguous user flows.

Thanks for visiting! I go into more detail on my work over calls/interviews. Feel free to reach me here.

But Wait, There's More!

But wait, there's more!