Northstar
What?
An AI learning platform that teaches users how to use AI effectively, critically, and responsibly.
For who?
Designed for both curious beginners and skeptical users seeking confidence and clarity in AI adoption.
My role
UX/UI design, from research and strategy to interaction design, testing, and visual design.
Project duration
3 weeks
Tools
Figma
Why This Matters
As AI tools become increasingly integrated into everyday workflows, more people are using them without fully understanding how they work of the implications of their use. From data privacy concerns to over-reliance on generated outputs, users are navigating a rapidly evolving landscape with limited guidance.
This project explores how an educational product can help users build practical skills while encouraging responsible and informed use.
What Users Need
A product that can support all user levels, from frequent AI users to skeptics.
To feel more confident and intentional in their AI use.
To build practical skills while engaging critically with AI-powered tools.
The Solution
I designed a personalized recommendation experience that helps readers discovery through mood-, genre-, and theme-based suggestions tailored to their reading history and preferences.
How I did it
Understanding AI
Competitive Analysis
The most common learning methods for AI are self-guided learning (trial and error), online learning platforms, and informal research through search engines. Comparing these tools revealed key gaps in the AI learning experience.
Key Theme
While information about AI is widely available, there are very few accessible solutions that combine practical application, critical thinking, and responsible use into one cohesive learning experience.
Research Plan
With these insights in mind, my goal for further research was to learn how people currently use AI and how they feel about it so I could better understand ways to empower them to feel more in control of AI technology.
Guiding Questions
Why do people currently use AI?
How do they improve their use/knowledge of AI?
What is their goal with AI usage?
What gaps exist in their AI usage?
What aspects of AI make them fearful of it?
Learning From AI Users and Critics
User interviews
User interviews were necessary for deepening my understanding of people's AI use and experiences. These discussions focused on the following topics:
Current AI usage/habits
Pain points and frustrations
Perceived benefits
Concerns and hesitations
Hopes for the future of AI
5 participants
I talked with 5 interview participants about their experiences and opinions regarding AI. The participants ranged from daily AI users to AI critics.
16 questions
Interviews went through 16 questions about participants' engagement, hopes, and criticisms about AI.
12 insight groups
These discussions reveals 12 affinity map insights, informing future ideation.
Interview Insights
Theme 1: AI is primarily used for productivity-related tasks, specifically for writing and information synthesis/summary.
Theme 2: Users' primary concern about AI is data privacy and how their inputs are used.
Theme 3: Some users fear that AI may reduce critical thinking and independence.
Theme 4: There is a general lack of understanding of how AI works and when to trust it.
User-informed personas
Defining the Problem
Problem Statement
Critics and regular users alike share concerns about data privacy, security, and skill atrophy. This leads to misuse, uncertainty, and hesitation in adoption.
How might we…
help users build practical AI skills while encouraging critical thinking and responsible use, empowering them to feel more in control of AI?
Designing a Learning Experience
Identifying core features
Based on research of online learning platforms, I defined key features to support learning as the following:
Guided lessons to teach foundational concepts
Interactive quizzes to reinforce understanding
Simulated exercises to practice real-world AI usage
Curates resources to stay informed on current developments
These features were designed to balance education and application, helping users learn by doing.
Mapping the experience
To structure the product, I create user flows outlining how users would interact with various parts of the product (i.e., onboarding, lesson completion, and resource access). This helped ensure that the experience was cohesive, intuitive, and scalable.
Bringing Ideas to Life
Low-fidelity wireframes
Referencing my user flows, I developed low-fidelity wireframes focused on three key areas:
Onboarding experience
Lesson flow and interactions
Navigation structure
These wireframes prioritized clarity and usability over visual design.
Usability testing
I conducted usability testing with the low-fidelity wireframes to evaluate the initial designs and perceived value of the product. Participants found the experience generally straightforward, but user feedback revealed insights that guided iterations.
Task 1
Participants complete the onboarding flow where users are introduced to the product, set their learning goals, and try a sample lesson to understand how the app works.
Feedback:
The onboarding process was generally straightforward and easy to complete
Some screens were text-heavy, causing diminishing engagement with the app
Participants need clearer validation/feedback on the sample lesson
Task 2
Participants find their next lesson and start learning.
Feedback:
The lesson flow was generally clear and easy to follow
Participants wanted more "hands-on" lessons that allow them to practice interactions with AI
Adding "steppers" would better inform users of what to expect in the lesson
Task 3
Participants navigate to the Resource and Profile pages to evaluate the content on each page.
Feedback:
Some participants wanted more visual content on the Resource page to reduce the cognitive load of reading through a lot of text
Participants wanted more personalized content and metrics on the Profile page
Establishing the Brand
Styling
I aimed to develop a visual identity to support a tone that is approachable, clear, and modern. This included defining typography, color palette, and UI components through a style tile.
Brand logo
In addition to the palette and typography, I created a logo that reflected clarity and trust while maintaining a modern and education-centered image.
Refining the Experience
High-fidelity wireframes
I translated the low-fidelity wireframes into high-fidelity designs applying the established brand system, with a focus on the following:
Strengthening visual hierarchy
Reducing cognitive load
Creating a consistent and intuitive interface
Hi-fi usability testing
I conducted another round of usability testing with both new and returning participants using the high-fidelity designs. The goals of this testing session were to validate the effectiveness of my iterations, evaluate app cohesion, and identify areas of improvement.
Task 1
The "user" has just downloaded the app with the goal of improving their workflow. Upon opening the app, they are prompted to complete the onboarding process.
Findings:
Participants found the onboarding screens to be informative but text-heavy
Several participants tried to select multiple learning goals (default is meant to be just one selection)
Overall, participants liked the simplicity of the onboarding and appreciated the progress markers
Task 2
Assuming the "user" has been using the app for a few days, they open the app to continue learning. After completing their lesson, the "user" wants to view more information about the unit they're on.
Findings:
Some participants clicked on the wrong card to continue learning, revealing a misalignment on between the user's mental model and the interface
Participant clicked on the entire "current unit" card instead of just the "unit summary text", demonstrating a need for larger touch-points
Participants enjoyed the answer validation and feedback
Task 3
The "user" wants to learn more about AI privacy. They need to navigate to the page where they could find that. The "user" wants to modify their learning goal. The need to navigate to the page where they could do that.
Findings:
Participants were able to complete both tasks with minimal confusion
Some participants found the resource page too plain and uniform
Participants liked the visual information presented on the profile page
Iterations
Iteration 1: Reducing text
During the onboarding tests, many participants mentioned that the screens were text-heavy. This caused frustration in the length of the onboarding process and presented a potential drop-off point. Reducing text at this stage allowed for a faster onboarding process as well as better readability and comprehension.
Iteration 2: Improving visual language
I wanted to improve the overall visual language of the interface to clarify clickable elements and add more elements to break up text. This change enhanced visual hierarchy and made scanning easier.
Iteration 3: Hierarchy
Many participants failed to click the intended card during the "learning task". This revealed information hierarchy issues on the dashboard and a failure to match user mental models. Resolving this issue created a more polished and user-friendly experience.
The Final Design
Introducing Northstar
The final product is an EdTech app that helps users
Learn how AI works
Build practical skills through interactive lessons
Understand how to use AI responsibly
Stay informed through curated resources.
The experience combines education, interaction, and real-world application to support both new and experienced users.
Define your learning path
Set a learning goal and receive a tailored learning journey based on how you want to use and understand AI.
Experience a lesson firsthand
Preview the interactive lesson format, practice responding to prompts, and receive feedback to reinforce learning.
Build skills
Complete hands-on lessons that help you develop practical AI knowledge while encouraging critical thinking and responsible usage.
Track growth and stay oriented
View your progress, upcoming topics, and learning objectives to better understand where you are in your journey and what comes next.
Stay informed
Explore curated articles and videos about emerging AI trends, ethics, and real-world applications to build informed perspectives on AI's impact.
Reflection
This project reinforced the importance of designing for both skill-building and mindset shifts. Teaching people how to use AI is only part of the challenge. Helping them use it thoughtfully requires clarity, trust, and intentional design.
If I were to take this project further, I would like to explore:
Expanding interactive simulations for deeper learning
Personalizing lesson plans based on user behavior
Integrating real-time AI feedback into exercises
