Sequence

Inspiration

One of our teammates grew up watching their dad, a mechanic, patiently teach people how to fix things. From their experience, we all realized how hard it still is to fix everyday things because most instructions don’t match what you actually see in front of you. Tradespeople often get overlooked, even though their knowledge is vital and too often invisible – because the work was hard to learn, and doing it wrong is incredibly dangerous. Sequence aims to make this more accessible for all and make trade jobs more accessible.

What it does

Sequence detects issues in real time, highlights the exact areas that need attention, and walks you through the fix with voice explanations.

How we built it

We used react native expo, webrtc, gemini, deepgram, elevenlabs

Challenges we ran into

Getting AR labeling to work

Accomplishments that we're proud of

Building a smooth user interface with animations

What we learned

How to get real-time feedback from an AI agent, and how to specialize it for a specific group of people

What's next for Sequence

Shipping it out to the world

Built With

• deepgram
• elevenlabs
• expo.io
• gemini
• react-native