Studio Project
2025
Designing Aero: an AI + AR flight assistant for seamless airport navigation.
Aero reimagines the airport journey with real-time navigation, smart alerts, and personalized travel tips. I designed end-to-end flows and high-fidelity prototypes for Aero, an AI- and AR-powered flight assistant.
My work included real-time navigation, context-aware notifications, and personalized travel tips—creating a system that reduces stress and makes airport journeys more intuitive.
Aero reimagines the airport journey with real-time navigation, smart alerts, and personalized travel tips. I designed end-to-end flows and high-fidelity prototypes for Aero, an AI- and AR-powered flight assistant.
My work included real-time navigation, context-aware notifications, and personalized travel tips—creating a system that reduces stress and makes airport journeys more intuitive.
Aero reimagines the airport journey with real-time navigation, smart alerts, and personalized travel tips. I designed end-to-end flows and high-fidelity prototypes for Aero, an AI- and AR-powered flight assistant.
My work included real-time navigation, context-aware notifications, and personalized travel tips—creating a system that reduces stress and makes airport journeys more intuitive.
Role
Product Designer
Team
Sole Member
Timeline
Jan – May, 2025
Introduction
Introduction
Introduction
Duolingo Video Call lets learners practice speaking in real time with Lily, a Duolingo character. While it enabled live interaction, it wasn’t fully optimized as a learning path. Learners often treated it like a casual video chat, and it lacked structured ways to track progress or support practice in a playful, effective way. Our goal was to reframe Video Call into a mission-driven learning experience.
We set out to design gamified, scalable session types that reduced stress while encouraging meaningful speaking practice — all within the constraints of existing architecture and backend prompting.
I was in charge of new UI and scaffolding of this session.
Duolingo Video Call lets learners practice speaking in real time with Lily, a Duolingo character. While it enabled live interaction, it wasn’t fully optimized as a learning path. Learners often treated it like a casual video chat, and it lacked structured ways to track progress or support practice in a playful, effective way. Our goal was to reframe Video Call into a mission-driven learning experience.
We set out to design gamified, scalable session types that reduced stress while encouraging meaningful speaking practice — all within the constraints of existing architecture and backend prompting.
I was in charge of new UI and scaffolding of this session.
Duolingo Video Call lets learners practice speaking in real time with Lily, a Duolingo character. While it enabled live interaction, it wasn’t fully optimized as a learning path. Learners often treated it like a casual video chat, and it lacked structured ways to track progress or support practice in a playful, effective way. Our goal was to reframe Video Call into a mission-driven learning experience.
We set out to design gamified, scalable session types that reduced stress while encouraging meaningful speaking practice — all within the constraints of existing architecture and backend prompting.
I was in charge of new UI and scaffolding of this session.
Challenges
Challenges
Challenges
How might we reframe Video Call into a mission-driven experience?
How might we adapt game mechanics into speaking practice?
How might we design scalable, low-stress session types?
How might we reframe Video Call into a mission-driven experience?
How might we adapt game mechanics into speaking practice?
How might we design scalable, low-stress session types?
How might we reframe Video Call into a mission-driven experience?
How might we adapt game mechanics into speaking practice?
How might we design scalable, low-stress session types?
Current Video Call
Current Video Call
Current Video Call
Currently, Video Call follows a very similar pattern [left], and can feel repetitive after several calls. We've been experimenting with new gamified and winnable sessions [right] in video call to add delight and reward metrics like words spoken
Currently, Video Call follows a very similar pattern [left], and can feel repetitive after several calls. We've been experimenting with new gamified and winnable sessions [right] in video call to add delight and reward metrics like words spoken
Currently, Video Call follows a very similar pattern [left], and can feel repetitive after several calls. We've been experimenting with new gamified and winnable sessions [right] in video call to add delight and reward metrics like words spoken
Role
Product Design Intern
Team
14 People
Timeline
Jun – Aug, 2025
Gamifying AI-powered video call experiences for stress-free speaking practice
Designed and prototyped new gamified Video Call session, creating user flows, wireframes, mockups, and interactive prototypes in Figma, Play, and Rive
Lead 2 design critique sessions, collaborated with learning designers to generate prompts for Video Call backend machine learning systems, iterated based on stakeholder feedback to balance user experience with engineering constraints
Designed new mechanic that will be used in future Video Call sessions.
What Am I Drawing
What Am I Drawing
What Am I Drawing
User research revealed learners often treated Video Call as a casual chat, which left speaking practice unfocused. We needed a way to add structure without losing the playful feel.
To solve this, we borrowed from I-Spy, a game with a familiar speaking pattern of asking and answering questions. In "What Am I Drawing", Lily sketches an object, and learners must ask questions and guess it in their target language. Learners can ask for hints, and if their guesses are getting closer, we used proximity feedback animation to let them know they're getting "closer" to the right answer.
We believed this evolution gives learners a clear goal, reinforces turn-taking with purpose, and makes the session feel more rewarding and replayable for Max subscribers
User research revealed learners often treated Video Call as a casual chat, which left speaking practice unfocused. We needed a way to add structure without losing the playful feel.
To solve this, we borrowed from I-Spy, a game with a familiar speaking pattern of asking and answering questions. In "What Am I Drawing", Lily sketches an object, and learners must ask questions and guess it in their target language. Learners can ask for hints, and if their guesses are getting closer, we used proximity feedback animation to let them know they're getting "closer" to the right answer.
We believed this evolution gives learners a clear goal, reinforces turn-taking with purpose, and makes the session feel more rewarding and replayable for Max subscribers
User research revealed learners often treated Video Call as a casual chat, which left speaking practice unfocused. We needed a way to add structure without losing the playful feel.
To solve this, we borrowed from I-Spy, a game with a familiar speaking pattern of asking and answering questions. In "What Am I Drawing", Lily sketches an object, and learners must ask questions and guess it in their target language. Learners can ask for hints, and if their guesses are getting closer, we used proximity feedback animation to let them know they're getting "closer" to the right answer.
We believed this evolution gives learners a clear goal, reinforces turn-taking with purpose, and makes the session feel more rewarding and replayable for Max subscribers
Window Collapse
Window Collapse
Window Collapse
Senior designers on the Video Call team had long wanted a way for Video Call to collapse into a picture-in-picture window, but it hadn’t yet been solved. I partnered with engineering to design and prototype the mechanic, working through constraints to make it lightweight and reusable.
This solution will now be implemented across future Video Call sessions, making it a core interaction pattern moving forward.
Senior designers on the Video Call team had long wanted a way for Video Call to collapse into a picture-in-picture window, but it hadn’t yet been solved. I partnered with engineering to design and prototype the mechanic, working through constraints to make it lightweight and reusable.
This solution will now be implemented across future Video Call sessions, making it a core interaction pattern moving forward.
Senior designers on the Video Call team had long wanted a way for Video Call to collapse into a picture-in-picture window, but it hadn’t yet been solved. I partnered with engineering to design and prototype the mechanic, working through constraints to make it lightweight and reusable.
This solution will now be implemented across future Video Call sessions, making it a core interaction pattern moving forward.
Dynamic FAB
Dynamic FAB
Dynamic FAB
We adapted a pattern from Duolingo Adventures into a dynamic floating action button that displayed the goal on opener, condensed into a timer during play, and transitioned into a winning state with celebratory feedback.
By consolidating the session’s key information into one component, the FAB made the experience more efficient and easier for learners to follow.
We adapted a pattern from Duolingo Adventures into a dynamic floating action button that displayed the goal on opener, condensed into a timer during play, and transitioned into a winning state with celebratory feedback.
By consolidating the session’s key information into one component, the FAB made the experience more efficient and easier for learners to follow.
We adapted a pattern from Duolingo Adventures into a dynamic floating action button that displayed the goal on opener, condensed into a timer during play, and transitioned into a winning state with celebratory feedback.
By consolidating the session’s key information into one component, the FAB made the experience more efficient and easier for learners to follow.
Connect to Content
Add layers or components to swipe between.
Conclusion
Conclusion
Conclusion
As a result of my work on Duolingo Video Call, I designed and prototyped What Am I Drawing — the first mission-based, gamified session type.
The prototype earned strong buy-in across teams and introduced a scalable mechanic that will now be used in future Video Call sessions, a result I’m really proud of.
As a result of my work on Duolingo Video Call, I designed and prototyped What Am I Drawing — the first mission-based, gamified session type.
The prototype earned strong buy-in across teams and introduced a scalable mechanic that will now be used in future Video Call sessions, a result I’m really proud of.
As a result of my work on Duolingo Video Call, I designed and prototyped What Am I Drawing — the first mission-based, gamified session type.
The prototype earned strong buy-in across teams and introduced a scalable mechanic that will now be used in future Video Call sessions, a result I’m really proud of.
*You can try this session in the app today if you’re in the beta, subscribed to Max, and enable session picker in the debugger menu, selecting "What Am I Drawing".
*You can try this session in the app today if you’re in the beta, subscribed to Max, and enable session picker in the debugger menu, selecting "What Am I Drawing".
*You can try this session in the app today if you’re in the beta, subscribed to Max, and enable session picker in the debugger menu, selecting "What Am I Drawing".