StockDigestAI

Inspiration

The inspiration for this project came from a desire to make stock investing more accessible to everyone. The financial world can be intimidating, with an overwhelming amount of information and fast-moving markets, which often leaves everyday investors feeling disconnected. I wanted to bridge this gap by leveraging AI to simplify the process of staying informed about stocks. By using Retrieval-Augmented Generation (RAG) to find the most relevant stock news articles in real-time, my goal was to provide investors with timely insights that empower them to make smarter decisions without needing to sift through countless sources. This project is driven by a belief that the power of the stock market should be within reach for all.

What it does

This project is an AI-powered web platform that provides users with real-time insights into the stock market by retrieving and analyzing the most relevant news articles. Using a technique called Retrieval-Augmented Generation (RAG), the platform continuously scrapes financial news, processes it to extract key information, directing users to articles that have the best insights. Users can quickly access the latest headlines and stock recommendations, making it easier to stay informed and make investment decisions without needing to manually research the market. The platform delivers an accessible, data-driven approach to stock investing by offering personalized, timely news analysis to help users navigate the complexities of the financial world.

How I built it

I developed this project with a React.js frontend and a Flask backend, creating a seamless user experience for accessing stock insights. For the AI-driven retrieval-augmented generation (RAG), I integrated a TiDB vector database, enabling fast and accurate retrieval of relevant stock news. Together, these technologies power the real-time analysis and insight generation, making financial data easily accessible to users.

Challenges I ran into

One of the major hurdles in this project was the compatibility issues with certain libraries in using a combination of Flask for the backend and React for the front end. Additionally, ensuring that the AI model retrieved and processed real-time stock news efficiently while maintaining accuracy required a great deal of fine-tuning. Balancing the complexity of Retrieval-Augmented Generation (RAG) with the need for a smooth user experience was a persistent challenge, but overcoming these obstacles ultimately strengthened the project.

Accomplishments that I'm proud of

I'm proud that I was able to implement an end to end Retrieval-Augmented Generation System in an idea that has real world applications in making stock trading more accessible to people.

What I learned

Through this project, I significantly enhanced my full stack development skills, deepening my understanding of both frontend and backend technologies. I also gained valuable experience working with advanced databases, particularly in integrating and optimizing a vector database for real-time retrieval.

What's next for StockDigestAI

I plan to publicly launch this project, making it accessible to a wider audience, while also expanding its capabilities by integrating diverse data sources. By incorporating additional types of data, I aim to provide users with even deeper insights and more comprehensive financial analysis, further enhancing the platform's value and utility. I also plan to make improvements to the front end of the application for a better user experience.

Built With

• javascript
• python
• react
• sentence-transformers
• tidb