Skip to Main Content

Our project is an intelligent finance assistant that connects to a user’s Gmail account, automatically extracts transaction details from emails, and generates personalized financial summaries. It uses Google AI Studio with the Gemini 2.5 Flash model to parse unstructured email text into structured data, such as merchant names, amounts, and dates. Through carefully designed prompts in AI Studio, Gemini interprets diverse email formats with high accuracy. The structured data is then analyzed to provide spending insights and answer natural language queries like “How much did I spend on travel last month?” using Gemini’s reasoning and natural language understanding capabilities. This project demonstrates how AI-powered prompt design and multimodal reasoning can transform everyday email data into actionable financial intelligence.

Team

hetvi4700smitra07Rithz7+1

Placement

Participant

More from Gemini Vibe Code Hackathon – SF

A local sandbox platform that lets product managers review and comment on their team’s code without touching the terminal. It spins up per-repo Docker…

www.loom.com/…

**NOTE: submitting the project again because of an error in loom video** Our project enables rapid prototyping in fashion…

www.loom.com/…

Youtube video: https://youtu.be/KSVpJFqF5hg Hi, I created mcp-game-asset-gen - an MCP server that allows Gemini CLI to easily generate 3d models and game…

MCP Game Asset Generator project preview

“FashionCollab is a rapid prototyping environment for design — except the prototype isn’t code, it’s clothing. Natural language drives iteration. The system…

www.loom.com/…

AI Schema Generator is an intelligent developer tool that turns any unstructured data like a receipt image, PDF form, spreadsheet, or text into a complete…

Team Poseidon project preview

We built a Cursor for Google Sheets. There are 900 million Google Sheets users each month, even more than the 700 million users Excel has. However, the…

SheetSense project preview