SignalX

AI powered market intelligence platform combining real time stock data macro signals and prediction market sentiment to generate instant actionable insights and shareable analysis

Comment

Inspiration

Most financial tools today show raw data but don’t help users make decisions. Traders and analysts often need to combine multiple sources like stock prices, macroeconomic data, and sentiment manually.

I wanted to build something that answers a simple question:

“What is the decision, not just the data?”

That’s how SignalX was born — a system that combines multiple signals and gives a single intelligent output.

What it does

SignalX is an AI-powered market intelligence platform that:

  • Takes a simple input like tesla or nvidia
  • Converts it into the correct stock symbol
  • Fetches real-time market data
  • Combines macroeconomic signals and prediction market sentiment
  • Outputs a clear decision: growth, decline, or neutral

It also allows users to share insights directly to X (Twitter) in one click.

How I built it

The system is built using a FastAPI backend deployed on Zerve AI platform and a modern React frontend.

Backend (Zerve)

  • Built using FastAPI
  • Deployed on Zerve cloud
  • Exposes a /predict API endpoint

The backend performs:

  1. Symbol Resolution
  • Converts user input (like “tesla”) → TSLA
  1. Market Data Fetching
  • Uses Stooq API for real-time stock prices
  1. Macro Data Integration
  • Uses FRED API to fetch interest rates
  1. Sentiment Analysis
  • Uses Polymarket API to extract probability-based sentiment
  1. Decision Engine
  • Combines all signals into a final prediction

Frontend

  • Built with React + TypeScript + Tailwind CSS

  • Users can:

    • Enter any stock or company name
    • View real-time analysis instantly
    • Share insights directly to X

Challenges I faced

1. API Failures in Deployment

Initially used Yahoo Finance and Alpha Vantage, but they failed in the cloud due to rate limits and restrictions.

Solution: Switched to Stooq API which is stable and has no limits.

2. Symbol Resolution Problem

Users entered names like “microsoft” instead of “MSFT”.

Solution: Built a hybrid resolver to convert natural input into valid stock symbols.

3. Data Consistency Issues

Some APIs returned empty or inconsistent responses.

Solution: Added fallback handling and structured response validation.

4. Frontend–Backend Integration

Handling CORS and ensuring smooth communication between UI and deployed API.

Solution: Enabled CORS middleware and standardized API response format.

What I learned

  • How to design a real-time data pipeline
  • Handling API failures in production environments
  • Building resilient backend systems
  • Combining multi-source intelligence into a single decision
  • Deploying AI systems using Zerve platform

What's next

  • Real-time streaming updates
  • Stock charts integration
  • AI-based explanation using LLMs
  • Portfolio-level intelligence

Final Thought

SignalX is not just a dashboard.

It is a decision intelligence system that transforms raw market data into actionable insights.

Built With

Share this project:

Updates