Skip to content

Q&A tool built on the latest Gemini generative AI, help investors, analysts, and executives interact with annual reports. By leveraging Retrieval Augmented Generation (RAG), it dynamically retrieves and generates precise, context-aware answers from extensive financial documents, saving hours of manual analysis and reducing oversight risks.

Notifications You must be signed in to change notification settings

andreanstev/langchain_gemini_qa_ar

Repository files navigation

Q&A from Annual Report

This project was presented as a participation at the Google AI Hackathon.

https://devpost.com/software/q-a-annual-report?ref_content=my-projects-tab&ref_feature=my_projects

Demo: https://colab.research.google.com/drive/1xrniSllXxTmqvXPDMmRTb-HNQylQEjp3?usp=sharing

Video: https://youtu.be/E23doQ9rd30?si=LUfNMgirjFNvRFK2

document_extraction.ipynb: The code loads a PDF, splits it into text chunks, and creates a vector representation of each chunk.

qna.ipynb: This code builds a system for having a conversation about the content in a PDF.

About

Q&A tool built on the latest Gemini generative AI, help investors, analysts, and executives interact with annual reports. By leveraging Retrieval Augmented Generation (RAG), it dynamically retrieves and generates precise, context-aware answers from extensive financial documents, saving hours of manual analysis and reducing oversight risks.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published