Bayse_terminal_backend is a backend system built during a "Build with AI" hackathon that serves as an African macroeconomic signal engine for prediction markets. The project tackles a genuine gap in financial tooling: while prediction markets have exploded in popularity globally, the infrastructure for monitoring and pricing events around African economies remains underdeveloped.

The system aims to do several things:

• Monitor African economic indicators — tracking macro signals like inflation rates, currency movements, and policy changes across the continent
• Detect anomalies — identifying when economic data deviates from expected patterns, which could signal market-moving events
• Auto-generate prediction market events — translating detected anomalies into structured market questions that traders can bet on
• AI-priced opening probabilities — using AI to set initial odds for newly created markets, solving the cold-start problem that plagues new prediction market questions

What makes this project interesting is its specificity. Rather than building yet another generic prediction market platform, the team focused on a concrete problem: African macroeconomic data is scattered, inconsistent, and underserved by existing financial platforms. By building an automated pipeline from raw economic signals to tradeable prediction markets, they're creating infrastructure that could make African economic events more legible to global participants.

The AI pricing component is particularly noteworthy. Setting fair opening probabilities for prediction markets is notoriously difficult — too wide and nobody trades, too narrow and the market maker gets arbitraged. Using an LLM to synthesize available information into an initial probability estimate is a clever application of generative AI to a quantitative problem.

This repo would be useful for developers working on prediction markets, fintech builders focused on African markets, hackathon participants looking for inspiration on combining AI with financial data pipelines, and anyone interested in the intersection of anomaly detection and market creation.