BIAZ Finance

Main page window.
Portfolio Analysis
Influence of News on Stocks.

Inspiration

Build a continuously active AI analyst that correlates news with market data — and learns from its own predictions over time.

What it does

Streams real-time + historical financial news and price data, analyzes impact with AI, predicts market direction, and continuously improves its strategy.

How we built it

Using Apify for live news, Kaggle for historical data, Redpanda for unified streaming, TrueFoundry for LLM analysis, and Senso for self-improvement feedback loops — all orchestrated in Node.js + Docker.

Challenges we ran into

Aligning news + price timelines, normalizing data, managing noise vs signal, ensuring real-time performance, and designing a meaningful learning loop connecting all API and platforms.

Accomplishments that we're proud of

Built a unified, backtest + live-capable streaming engine, ran LLM-based market reasoning, and implemented self-improving feedback logic — all within hackathon time.

What we learned

Data quality matters more than model. Real autonomy needs feedback, not just LLMs. Streaming infra (like Redpanda) is a game-changer vs. traditional APIs.