Kin — Your AI Companion

Meera in action: A friendly, multilingual AI companion for everyday conversations.
Everyday support: Helping older adults manage daily routines and medication reminders.
Privacy by choice: Giving users control over personal memories and caregiver sharing.

Inspiration

Kin was inspired by the idea that technology should make people feel more connected, not more dependent on complicated devices. Many older adults may experience loneliness, forget daily tasks, or find modern applications difficult to navigate. We wanted to create an AI companion that feels like a friendly conversation rather than another tool to operate. This led to Meera, the caring AI personality at the heart of Kin.

What it does

Kin is a voice-first AI companion designed to support older adults in their everyday lives. Its goal is to make technology feel more natural, accessible, and personal.

AI companionship: Provides a friendly space for everyday conversations.
Multilingual interaction: Aims to support communication in English, Kannada, and Hindi.
Daily reminders: Helps users keep track of routines, appointments, and medication schedules.
Privacy-conscious memory: Gives users control over what information stays private and what can be shared with trusted caregivers.
Safety-oriented support: Is designed to help users find guidance and connect with trusted people when they need assistance.

How we built it

We designed Kin as a web application with a simple, approachable interface and a backend that connects to Google's Gemini API for conversational AI. Browser-based speech recognition helps users interact through voice, while the application brings together conversation, reminders, and privacy-focused memory concepts. We focused on keeping the experience straightforward so users would not need to navigate complicated menus to interact with their companion.

Challenges we ran into

One of our biggest challenges was connecting the frontend, backend, AI service, and deployment environment into one application. Configuring API access, resolving server connection issues, and making voice input useful across different languages required careful troubleshooting. Another important challenge was thinking about privacy: a companion may need to remember useful information, but users should remain in control of what it remembers and what others can access.

Accomplishments that we're proud of

We are proud of turning a human-centred idea into a concrete application concept rather than building another generic chatbot. Kin brings together companionship, voice interaction, everyday assistance, and privacy-conscious memory in one experience. We are especially proud of the principle behind the project: older adults should be able to receive support without giving up their independence or control over personal information.

What we learned

We learned that building useful AI involves much more than connecting a model to a user interface. Accessibility, reliability, privacy, and trust are equally important. We also learned how challenging it can be to integrate multiple technologies into a working application and how important it is to test features in realistic conditions instead of assuming they work simply because they have been implemented.

What's next for Kin — Your AI Companion

Our next goal is to make Kin more reliable and useful in everyday life. We want to improve multilingual voice recognition, build dependable reminder functionality, strengthen user-controlled memory permissions, and develop carefully designed ways to contact trusted caregivers during appropriate situations. We also want to test Kin with real users and incorporate their feedback to make the experience more accessible and natural.

Our long-term vision is simple: Kin should help people feel heard, supported, and connected while keeping their independence and privacy at the centre.