-
-
Home Page - WhaleWatcher
Inspiration
As a CS and Maths duo, we saw that institutional "Whale" data is gatekept by expensive terminals. We wanted to build a "Machine of Loving Grace" that decompiles this power into Financial Literacy for retail savers in emerging markets.
What it does
WhaleWatcher is a live tutor that connects to exchange WebSockets. It uses Claude to "translate" high-velocity trade data into real-time, human-readable lessons on market microstructure and institutional tactics.
How we built it
Data Engine: High-performance listener for raw JSON trade streams. AI Reasoning: Claude 3.5 Sonnet Math Heuristics: Identifying significance when trade volume V > \sigma{15s} UI: A triple-pane dashboard.
Challenges we ran into
Managing Data Velocity vs. Reasoning Latency. Trying to implement a Significance Filter to ensure Claude only analyzed the most impactful moves, preventing API rate limits.
Accomplishments that we're proud of
We built a working prototype that turns a "Black Box" into a transparent classroom.
What we learned
We also refined our ability to build high-stakes data pipelines under extreme time pressure.
What's next for WhaleWatcher
Integrating Proofs to verify the data's integrity and expanding to multi-chain feeds to provide for all decentralized markets.
Log in or sign up for Devpost to join the conversation.