Stock Intelligent Advisor

Our home page
chat page

Inspiration

The motivation behind our project stems from the recognition that gathering financial data on publicly traded companies is an exhaustive and intricate process. Our objective is to develop a system designed to empower investors, particularly individual ones, by centralizing financial and pertinent information from diverse companies. This platform is envisioned to not only consolidate data but also interpret specific inquiries, delivering tailored responses to enhance decision-making and insight.

What it does

Our project introduces a pioneering system dedicated to streamlining the process of accessing and analyzing financial data for publicly listed companies. Aimed at demystifying financial information for a broad audience, especially individual investors, this platform aggregates financial statements and relevant company information into a user-friendly interface. By leveraging natural language processing capabilities, it allows users to pose specific questions and receive precise, comprehensible answers. This not only facilitates a deeper understanding of a company's financial health but also aids in making informed investment decisions by providing a holistic view of financial metrics and trends at the touch of a button.

How we built it

Our platform utilizes a Browser-Server (BS) architectural framework, with our development team divided into two specialized groups dedicated to front-end and back-end development respectively. For the back-end, we've chosen Flask, a Python-based web framework, to facilitate the creation of RESTful APIs, which are essential for our system's operation. These APIs interact with advanced technologies such as Retrieval-Augmented Generation (RAG), Large Language Models (LLM), and Vector Databases (Vector DB) for efficient storage and retrieval of financial data. On the front-end, we employ ReactJS to develop a user-friendly interface that displays company reports and enables users to communicate with our chatbot using natural language. This division of labor and choice of technologies ensures our system is both powerful in data processing and user-friendly, offering a seamless experience for accessing and analyzing financial information.

Challenges we ran into

Given that our team is working with Retrieval-Augmented Generation (RAG) and Vector Databases (Vector DB) for the first time, mastering these technologies within one and a half days presents a significant challenge. This is particularly true as half of our team is dedicated to focusing on front-end development, limiting the resources available to delve deeply into these complex systems.

Accomplishments that we're proud of

Initially, our team was able to create the AI prototype in under two days, covering the stages of ideation, designing the framework, executing the plan, and crafting the presentation. Subsequently, the entire team worked together flawlessly, intelligently distributing tasks among members to ensure efficient cooperation and mutual support.

What we learned

In the process of developing this project, our initial focus was on understanding the technical aspects of Retrieval-Augmented Generation (RAG, specifically LlamaIndex) and Vector Databases. However, beyond the technical learning, we significantly enhanced our teamwork skills, collaborating effectively to reach a common objective.