Inspiration
Being incredibly annoyed at awful email applications and being inundated by BRICKED inboxes
What it does
Our mobile application is an AI powered mailbox assistant. Our assistant can rank emails, plucks the most important semantic information from your emails such as key tasks and deadlines. We even allow users to be able to set their own preferences for how to rank their emails!
How we built it
We used the newest AI LLM to date, Claude-3, pushing the boundaries of what is currently possible with Generative AI technology. We built a mobile app using React Native to ensure the app was as relevant as possible to our usecase.
Challenges we ran into
Using Claude-3, the most powerful LLM at the time of this competition, was definitely one of the most challenging parts of this project. We had to find ways to efficiently deal with not only speed due to high demand, but rate limiting. And despite this, Claude-3 still has its limitations. We tackled this by reading deep into documentation and even research papers for prompt engineering recommendations, and utilising advanced techniques like Chain of Density (CoD) prompting, and prompt rewriting.
Accomplishments that we're proud of
We are most proud of the fact that we were able to successfully build a mobile application despite neither of us having any experience making one.
What we learned
How to make a mobile app and experiencing using an Anthropic LLM.
What's next for BRICK@ed.up
BRICK@ed.up WILL return for world supremacy. We will not be reading any replies as we know we are correct.
Built With
- claude-3
- expo.io
- generative-ai
- llm
- react
- react-native
- socket.io
- typescript
Log in or sign up for Devpost to join the conversation.