Stock Saga: A Gemini Integrated Financial Trading Simulator

Intraday traders are severely disadvantaged by speed. In our experience, one of the key reasons they slippage across trades is that their workflows are disjointed. A typical retail setup requires the trader to switch between a charting website, a separate broker terminal, and a social media news feed. Every click between tabs costs money. We propose an engineering solution, Stock Saga, which is a proprietary web platform that we built to solve the problem to bringing dis- parate trading tools together into a single high-velocity interface. In order to provide a real-time, financial data processing application without interface lock- ups, we designed the backend service architecture away from the conventional centralized monolithic servers and used a highly concurrent JAMstack archi- tecture on distributed Deno edge nodes. We also add a custom Convolutional Neural Network (CNN) integration in the charting DOM to detect trading pat- terns automatically on the client side, paired with Language Model inference pipeline to score breaking news. Before risking capital, users validate their algo- rithmic concept with our mathematical backtesting engine. We also present a custom Know Your Customer (KYC) pipeline that uses facial geometry vectors to verify identity documents without manual review. Our empirical stress tests show that this edge-routed architecture reduces API response latency by 71.9% compared to traditional centralized architectures and maintains browser mem- ory footprints under high-frequency WebSocket updates. In the end, this work presents a professional-grade system to backtest and run traders’ strategies with zero UI latency.