Our Inspiration
We've all watched it happen, a headline drops, a stock moves, and by the time anyone figures out why, the moment has already passed. We're a team that cares deeply about making sense of complex information, and when we saw the challenge of turning financial text into reliable decision signals, it clicked immediately. The news that moves markets is out there, it's just buried across dozens of outlets, filtered through noise, faster than any analyst can read. Dynamism came from that frustration: we wanted to build something that reads the way a sharp analyst would, that shows its work, and that gives human judgment a fighting chance against the volume — not a black box predicting the future, but a transparent system that says here are the narratives, here is the evidence, you decide.
What it does
- Dynamism takes a user query and then scrapes the web for related links. We then run these links through an embedding model to get vector weights, which we pass through a segmentation model to find semantic similarities, before ultimately pushing those into our 768 dimensional vector database.
- From there we run a parametric U-map implementation to map our 768 dimensional latent space into a 3 dimension graph while maintaining 3 dimensional clusters in order to then present this to the user in the front end.
How we built it
- We used ChatGPT to bounce ideas around.
- Then we made flow chart of our project.
- Then we mocked up the UI in Figma.
- Then we started writing the back end and front end making use of agentic tools including Google Antigravity.
Challenges we ran into
- Time Crunch
- Debating on whether to use Rust or Typescript and ultimately choosing Rust
- Linking our logic between Rust to Typescript
- Learning libraries in a short amount of time.
- Implementing Gemini API
Built With
- fastembed
- geminiapi
- lancedb
- react
- rust
- spacetimedb
- typescript
- vite
Log in or sign up for Devpost to join the conversation.