Inspiration
Managing personal finances can be overwhelming and time-consuming. We wanted to build an AI-powered assistant that could understand natural language queries, provide meaningful insights from transaction data, and help people make better financial decisions — just like talking to a personal finance analyst.
What it does
The Personal AI Finance Assistant allows users to:
Input and save their transactions with description and AI categorizes them.
Ask questions like "Where am I overspending?", "How much did I spend on groceries last month?", or "Show my income trends".
Get personalized answers and visualizations using AI + data analysis.
It uses Google Gemini to generate Python code for analyzing the data.
How we built it:
Backend AI: Gemini (via Google Generative AI) is used to understand user queries and generate context-aware analysis code.
Agent Communication: Uses uAgents framework with mailbox routing to allow users and agents to send messages and receive AI-generated insights.
Data Visualization: Leveraged Plotly Express for clean charts.
Challenges we ran into
Managing AI-generated code safely and executing it with proper error handling.
Ensuring smooth interaction between user agents and AI agents over the uAgents framework.
Dynamically handling a wide range of queries and inconsistent data formats from user-uploaded Excel files.
Making proper prompts.
Accomplishments that we're proud of
Created an end-to-end system that lets a user talk to their data like chatting with an assistant.
Used LLM-generated Python code to dynamically analyze real data with context.
Learned and integrated new tools like uagents, AgentVerse and Google Generative AI API in a short time.
What we learned
How to build and deploy AI agents using the uAgents framework.
How to guide LLMs to generate safe, executable, and insightful analysis code.
What's next for Personal AI Finance Assistant
Real-time transaction syncing via APIs from banks or finance apps.
Improved intent detection using custom fine-tuned models for financial domains.
Personal budgeting advice with proactive tips based on spending trends.
Voice integration so users can talk to their assistant naturally.
Security enhancements for safe handling of personal financial data.
Log in or sign up for Devpost to join the conversation.