About Xchange
Team Members: Sujal Shah, Binayak Jha, Sachin Shrestha
We built Xchange because many people turn to social media to see what users are saying about the markets — to get investment ideas, understand sentiment, find options flows, or simply follow trending stocks.
But as the volume of financial chatter grows, so does the noise. It's hard to separate what’s relevant from what’s not. We often find ourselves lost in an endless feed of opinions without knowing which ones truly move the market.
Xchange was created to turn market chatter into market insight. It’s a social, data-driven platform inspired by modern social feeds, but focused entirely on finance. The idea is to help traders and investors understand what’s trending, how people feel about it, and how that collective sentiment might actually influence market moves.
⚙️ How We Built It
We used React + TypeScript to build a modular, production-ready frontend.
All state management is handled through React Context, and we used mock APIs for now to simulate live data streams from social media, Reddit, and financial sources. The architecture is fully designed to integrate with real APIs, including key social media APIs and AI APIs, with minimal refactoring.
The layout takes inspiration from familiar feed designs but reimagined for finance:
- Watchlist Manager to follow stocks of interest
- Sentiment Sidebar showing live bullish/bearish sentiment
- Unusual Flows section highlighting large or odd trades happening in real-time
- Paper Trading area for simulating positions based on option flows and social signals
We used TailwindCSS for clean, responsive styling and Framer Motion for smooth transitions and animations — ensuring that every user interaction feels natural and fluid.
🤖 The Role of AI
AI is the secret sauce powering our project’s real-time market sentiment insights—turning the chaos of social conversations into actionable, dynamic intelligence for traders and investors.
At the heart of our system is a large language model, leveraging its advanced capabilities to decode the emotional pulse of the market.
Instead of old-school formulas and rigid statistical models, we tap directly into modern AI’s language understanding. Every time a stock is mentioned, our AI delivers a rich JSON readout: overall, positive, negative, and neutral sentiment percentages, crafted in real time from the raw conversation flow.
No hand-tuned parameters. No clunky embeddings.
Just pure AI intuition—translating public mood into instantly accessible, structured numbers.
If—on the rare occasion—AI isn’t available, we seamlessly switch to a fast post-counting engine that keeps the sentiment ticker moving, ensuring zero downtime for our live dashboards.
All sentiment percentages are aggregated and streamed to users as slick, real-time visuals—so you never miss what the market is feeling, second by second.
This isn’t your average sentiment widget. It’s a live, AI-powered market mood ring, built for the next generation of fintech.
Let our AI be your edge—and see how machine learning transforms noise into opportunity, right as it happens!
💡 What We Learned
Through Xchange, we learned how complex it is to merge social data, financial information, and AI-driven insights into one cohesive product. Designing something that feels familiar — like a social platform — while keeping it data-driven and focused on clarity required thoughtful UI/UX planning.
We deepened our understanding of:
- Structuring React applications for real-time, scalable updates.
- Using AI embeddings for sentiment classification.
- Designing data visualizations that stay informative but intuitive.
- Bridging the gap between qualitative chatter and quantitative signals.
This project showed us that social sentiment isn’t just noise — it’s a measurable market force when analyzed correctly.
🚧 Challenges We Faced
- Creating a social media-like interface that feels seamless and modern while tailored for finance.
- Managing complex, animated state updates efficiently.
- Making mock data behave realistically in real time.
- Integrating AI sentiment simulation without an active backend.
- Ensuring full responsiveness and performance across devices.
🚀 What’s Next
We plan to connect Xchange with live social and financial APIs, pulling in sentiment data from social media and other finance communities.
With integrations to modern AI providers, we aim to scale our AI layer to continuously analyze real posts and track how discussions shape stock movements.
Ultimately, our goal is to make Xchange the go-to social-financial platform — where traders don’t just follow the markets, but understand the conversation driving them.
Built With
- github
- javascript
- react
- sql
- tailwind
- typescript
- vite
- x
- xai

Log in or sign up for Devpost to join the conversation.