Living.fm

Living.fm was born from our team's desire to be more intentional about how we spend free time at home. While routines shape who we are, many of our habits—like endless scrolling or streaming—are driven by tech designed to keep us hooked. We explored how self-awareness could act as gentle friction to interrupt these patterns. The result was Liv, a portable AI narrator that pairs ambient music with intermittent, slightly sassy, contextual commentary.
Role
Physical Computing, Prototyping, User Testing, LVM's, LLM's, Voice AI
Course
Designing with AI
4 days, 2024
Team
Juan Ignacio Garro
Wayne Kim
Instructors
Bjørn Karmann
Matteo Loglio

Catch began as an online payment method.

Catch rewards shoppers for using ACH or debit at checkout. Merchants save on credit card fees and offer store credit to boost loyalty, while shoppers earn high rewards (typically 10%) without the drawbacks of credit.
Description
Description

Adding new brands required checkout integration.

By 2024, we had integrated with around 70 merchants, but growth was limited by the need for merchant integration. Potential partner brands had scarce resources for integration, and due to our small network, we didn't yet have the traffic to demonstrate short term value — a real chicken-and-egg situation.

Rethinking routine

We began with a simple observation: many of our daily habits at home—like doomscrolling or binge-watching—are shaped by tech designed to capture our attention. We wondered if a little friction, in the form of self-awareness, could help shift those patterns toward something more intentional.

Designing Living.fm

We created Living.fm, a speculative product featuring Liv—a portable home narrator that pairs ambient music with fun, contextual commentary. Liv responds to the your behaviors, offering encouragement for certain activities, like cooking or stretching, or lightly roasting you for less desirable ones, like doom-scrolling.

Building Liv’s personality

Using Cursor, ElevenLabs, and Google Teachable Machine, we vibe-crafted Liv’s awareness and voice. She recognizes simple physical activities through a webcam and responds with commentary—sometimes helpful, sometimes sassy, always present. While the original plan was to have her comment on any type of activity, consistency issues led us to stick with several specific ones for the purpose of the prototype.
Bodystorming while narrating each other's activities helped us feel out timing and tone, and select a set of sample activities.
We used Google's Teachable Machines to train a machine learning model that recognizes our sample movements.

Considering privacy and control

While Liv's watchful eyeball intentionally announces her nosey nature, it's also a privacy feature. Turning her it downward opens the camera circuit, putting her to bed. Additionally, her AI models runs entirely on-device, so everything is run locally.

Iterating through feedback

Our early prototypes focused on interrupting doomscrolling, but user testing showed this just felt like nagging: "It's like an alarm. I just want to turn it off," one tester complained. So, we expanded Liv’s role to include positive reinforcement and ambient music, and in the future, plan to give users more control over when and how she speaks.
We iterated on Liv's form with sketching and lofi prototyping. The "narrator" metaphor was central to our explorations, so we wanted her form to reflect both watching and speaking.
When testing out different lofi prototypes with classmates, the general response to the eyeball was, "it's so creepy, it's cute," while the megaphone alluded to announcements, but didn't feel as personal. We decided to combine the two concepts.

A speculative exploration

Living.fm is not a product, but a provocation. Through this concept, we explored how AI, sound, and physical design might come together to make our home lives more mindful—one small interaction at a time.

Up next