CellMatch

Inspiration

Buying a phone should be an exciting experience. You're scrolling, looking for your new companion for what might be years. For many instead it's a confusing, extremely technical and stressing experience. finding the right phone often requires knowledge the average customer experiencing the technological divide might not be able to do. Our inspiration came from a single idea: What if finding your next phone was as easy as having a conversation? instead of having to work through what RAM, processor and camera to get, we wanted to build something that any customer can understand and get what matters to them.

What it does

Cellmatch is an AI-powered voice-first phone shopping assistant that helps customers discover the right smartphone through natural conversation. Rather than scrolling through endless product listings or completing long questionnaires, users can simply talk about their needs.

How we built it

We built Voicewatch using a modern web application architecture that combines conversational AI, voice technology, and personalized recommendation logic. Our frontend was developed using Next.js, React, and TypeScript, providing an interactive interface where users can start voice conversations and explore recommended devices. For voice interactions, we integrated ElevenLabs Conversational AI, enabling natural spoken conversations Behind the scenes, our conversational AI system processes user responses to extract structured preferences, including budget, camera quality, battery life, performance, and everyday usage.

Challenges we ran into

One of our biggest challenges was connecting real-time voice conversations with a structured recommendation system. Traditional chatbots can generate responses freely, but our application needed to translate natural language into specific preferences that our recommendation engine could understand.

Accomplishments that we're proud of

We're especially proud of turning a complicated shopping process into a conversational experience designed around accessibility. Rather than building another product comparison website, we approached the problem from the perspective of people who might not know where to begin. Some of our biggest accomplishments include:

Developing a voice-first shopping experience centered on natural conversation.
Integrating ElevenLabs into our custom application architecture.
Connecting conversational AI with personalized device recommendation logic. ## What we learned Building Voicewatch taught us that creating an effective AI application involves much more than connecting to a language model. We learned how to design systems where voice interfaces, conversational reasoning, structured data, and recommendation algorithms work together.

What's next for CellMatch

Our long-term vision is to turn Voicewatch into a complete AI-powered shopping companion that makes purchasing technology accessible to everyone. We'd want to also add more languages to make it accessible to even more communities.