INFOGO

InfoGo is a Retrieval Augmented Generation (RAG)-based application designed to enhance stock analysis by leveraging information from multiple online sources. The system gathers data from a diverse set of websites, including financial news, market analysis platforms, and expert blogs, to compile a comprehensive view of a company's stock status.

This intelligent assistant leverages RAG technology to intelligently retrieve relevant data, analyze it, and generate informed insights, helping investors make better decisions. By integrating information from multiple sources, the analysis is not limited to a single perspective, increasing the accuracy and reliability of stock predictions.

InfoGo is built with a modern tech stack that combines multiple programming languages and technologies to deliver a comprehensive financial analysis tool:

• Frontend: TypeScript with Tailwind CSS for a responsive user interface
• Backend: Python for RAG implementation and API services
• Additional services: C# (.NET) for data processing
• Data storage: MongoDB for flexible document storage
• Cloud infrastructure: AWS services for deployment and scaling
• AI: OpenAI and LangSmith integration for natural language processing
• Web scraping with Python to gather diverse financial data

InfoGo was developed as a collaborative project for HackUTD 2024, sponsored by Infosys. As part of the development team alongside Sujin and Yousuf, I contributed to the application architecture and implementation.

The development process involved:

• Designing a RAG-based architecture to efficiently retrieve and analyze financial data
• Implementing web scraping capabilities to gather information from multiple sources
• Developing natural language processing capabilities for generating insights
• Creating a user-friendly interface for accessing financial analysis
• Integrating with OpenAI and LangSmith for enhanced performance