Skip to content

FinTrackAI is a smart, secure, and interactive AI-powered assistant that helps users extract insights from their financial documents through natural language queries.

Notifications You must be signed in to change notification settings

Manraj29/FinTrackAI

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

3 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

FinTrackAI β€” Your Secure Financial Chat Assistant

FinTrackAI is a smart, secure, and interactive AI-powered assistant that helps users extract insights from their financial documents through natural language queries. Upload PDFs like bank statements, invoices, or salary slips, and chat with your data to uncover actionable insights β€” all while keeping your information encrypted and confidential.

Built with React, Node.js, Firebase, Google Gemini API, and Tailored Encryption Mechanisms.


πŸš€ Features

πŸ—¨οΈ Chat Interface

  • Create, rename, and delete multiple chat sessions.
  • Real-time AI answers via Gemini.
  • Store and sync messages to Firebase.

πŸ“Ž PDF Upload + Processing

  • Upload PDF files (e.g. bank statements, invoices).
  • Securely parsed and encrypted before backend processing.
  • Extracted text passed to Gemini for insights.

🧠 AI Financial Insights

  • Ask questions like:
    • "What’s my total expenditure this month?"
    • "Summarize my recent transactions"
    • "Are there any unusual expenses?"
  • Get smart summaries and analytics on your data.

πŸ” Data Security

  • Files never stored as-is.
  • Financial text is hashed + encrypted before backend processing.
  • Decryption occurs only during Gemini processing.
  • End-to-end data confidentiality assured.

πŸ“ Chat Management

  • Rename chats (e.g. β€œJan 2024 Bank Summary”).
  • Get AI-generated summaries via modal popup.
  • Sort chats based on recent activity.

🧰 Tech Stack

Layer Tech Used
Frontend React, TailwindCSS, Heroicons
Backend Node.js, Express
AI Gemini API (Google Generative AI)
Database Firebase Realtime DB & Firestore
Storage/Security In-memory PDF parsing, crypto, AES Encryption

πŸ§ͺ Example Prompts

After uploading a bank statement PDF:

  • "How much did I spend on food in March?"
  • "Give me a summary of my monthly salary deposits"
  • "Are there any duplicate transactions?"
  • "Generate a spending report with insights"
  • "What were the top 3 categories I spent on?"

Screenshots

Signup

Signup Screenshot

Login

Login Screenshot

Home Page / Chat Dashboard

Home Page Screenshot

First Prompt

First Prompt Screenshot

FinTrackAI Response

FinTrackAI Response Screenshot

Continuation of Chats

Chat Continuation Screenshot

Chat Options

Chat Options Screenshot

Summary of Chat

Chat Summary Screenshot

Deleting a Chat / Reset Password

Delete Chat or Reset Password Screenshot

Firestore

Firestore Screenshot

Example Statement used

Example img

Overview

How FinTrackAI Works?


FinTrackAI allows users to securely upload financial documents (PDFs) such as bank statements or invoices, which are parsed and encrypted on the backend before being analyzed by Google's Gemini Pro AI model. Users can then chat naturally with the assistant to extract personalized insights like expense summaries, unusual transactions, or monthly reports. All chats and document data are stored securely using Firebase, and the system ensures end-to-end confidentiality with encryption and hashing. The frontend offers a smooth experience with resizable chat sidebar, real-time AI messaging, PDF preview, and features like renaming, deleting, and summarizing chat sessions.

Disclaimer & Terms

FinTrackAI does not provide financial or legal advice.

  • All data is processed securely and never stored in raw form.
  • Use at your own risk β€” insights are AI-generated and may not be 100% accurate.
  • Do not upload highly sensitive documents unless you're sure of the environment.

About

FinTrackAI is a smart, secure, and interactive AI-powered assistant that helps users extract insights from their financial documents through natural language queries.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages