This project introduces an end-to-end trading system that leverages Large Language Models (LLMs) for real-time market sentiment analysis and trading signal generation. By synthesizing financial news and social-media streams, the system integrates sentiment-driven insights with technical indicators to support actionable investment decisions. FinGPT serves as the core sentiment model, ensuring domain-specific accuracy, while additional models validate performance across financial contexts. The framework combines live multi-source data processing, LLM-based summarization, and real-world strategy testing, deployed on Kubernetes for scalable operation. This comprehensive approach demonstrates how advanced LLM techniques can enhance trading strategies through timely and nuanced sentiment understanding.