Inspiration
Markets move on emotion long before fundamentals change. We built xSent to test whether real-time sentiment on X (Twitter) can forecast market movement transforming social emotion into actionable Kalshi trading signals through xAI.
How we built it
We combined three real-world APIs: X API: to fetch trending tweets for any topic. xAI: to analyze sentiment and assign polarity scores. Kalshi API: to retrieve live markets and simulate or place orders. Our FastAPI backend handles data collection, sentiment scoring and secure Kalshi authentication using asymmetric key signatures. The Streamlit frontend visualizes results showing sentiment distributions, confidence gauges and live Kalshi markets tied to each topic.
Challenges we ran into
X API rate limits was the biggest challenge and halted progress. Kalshi domain access required custom endpoint routing.
What we learned
Integrating multiple APIs securely and handling rate limits. Turning text sentiment into numeric trading signals. Visualizing emotion-to-market relationships in real time.
Log in or sign up for Devpost to join the conversation.