firstName

lastName

email

message

updatesConsent

Let’s talk... 💬

Personal

SaveDollar is a web-based personal finance assistant I built to manage and analyze my own expenses more intelligently. The app integrates AI-powered forecasting with financial dashboards to help users track their income, expenses, budgets, savings goals, and spending trends with ease.                                                      

# 💸 SaveDollar – Intelligent Personal Finance Forecasting App

**SaveDollar** is a smart personal finance assistant designed to help users make sense of their spending, categorize transactions, and predict future expenses. Built as a capstone-style solo project, SaveDollar showcases the intersection of **machine learning**, **data science**, and **frontend engineering**. It reflects my core strength as a data-driven developer who cares about delivering **user-centered insights** with simplicity and automation.

> “Everybody should learn to program a computer, because it teaches you how to think.” – Steve Jobs

## 🧠 Project Background

The idea for SaveDollar came from a real-world observation: most people, especially students and young professionals, struggle with managing their finances. While numerous apps exist, they often lack intelligent predictions and customizable insights. As a data science student passionate about **AI-driven solutions**, I wanted to create an app that not only tracks expenses but also **learns from your past behavior to forecast your future spending** — like a personal budgeting assistant that gets smarter with time.

SaveDollar was developed using a full-stack approach. The **frontend** was built using React, Tailwind CSS, and Shadcn/UI for a sleek and responsive interface, while the **backend model** was prototyped in Python using Scikit-learn to train a forecasting pipeline. The goal was to use historical transactions (imported as `.json`) and deliver dynamic forecasts with minimal user setup.

***

## ✨ Key Features

*   🔄 **Transaction Import**: Upload your personal `.json` file of bank transactions.
*   🧠 **Automated Categorization**: Transactions are sorted into standard categories like groceries, rent, transport, etc.
*   📊 **Interactive Dashboard**: Visualize spending trends across months and categories.
*   🔮 **ML-Based Forecasting**: Scikit-learn pipeline predicts your next month’s expenses by analyzing past transaction patterns.
*   🧩 **Component-Based UI**: React and Shadcn were used to build reusable, responsive, and accessible components.
*   🌙 **Dark Mode Support**: Optimized for late-night financial planning!

***

## 🔧 Tech Stack

| Layer        | Technology                     | Purpose                                                     |
| ------------ | ------------------------------ | ----------------------------------------------------------- |
| Frontend     | React + Vite + TypeScript      | Fast UI, modular component development                      |
| Styling      | Tailwind CSS + Shadcn UI       | Custom styling with utility classes and prebuilt components |
| Backend (ML) | Python + Scikit-learn + Pandas | Forecasting future expenses using regression                |
| Build Tool   | Vite                           | Lightning-fast development and production build             |
| Deployment   | GitHub Pages / Vercel          | Hosting the web app                                         |
| File Parsing | JSON import logic in JS        | Enables importing user transactions                         |

***

SaveDollar expects a `.json` file structured like this:

## 🧠 Why This Project Matters

Most budgeting apps stop at *what* you spent. SaveDollar tells you *what you might spend next.*

In a world where AI recommends your next song, movie, or email — why not your next financial decision?

This project reflects my commitment to applying data science in ways that are **smart**, **ethical**, and **human-focused**. It’s not about complexity — it’s about clarity.

## 🔗 Links

*   🔗 **Live Demo**: [Save$](https://infinitudelogistics.in/)

*   📁 **GitHub Repo**: [Repo](https://github.com/HarshDW09)

*   🌐 **Portfolio**: [Portfolio](https://harshd09.netlify.app/)

*   💼  [LinkedIn](https://www.linkedin.com/in/harshdwivedi050/)


## 👨‍💻 Author

**\[Harsh Dwivedi]**

🎓 Data Science Graduate | 🧠 Machine Learning Enthusiast | 💻 Frontend Explorer



DEMO

SaveDollar

It’s hard to imagine that I’ve that I wrote all this code by myself, probably because I worked with an entire team :) but they definitely followed my lead most of the time.

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Sed ante lorem, tincidunt ac leo efficitur, feugiat tempor odio. Curabitur at auctor sapien. Etiam at cursus enim. Suspendisse sed augue tortor. Nunc eu magna vitae lorem pellentesque fermentum. Sed in facilisis dui. Nulla molestie risus in mi dapibus, eget porta lorem semper. Donec sed facilisis nibh. Curabitur eget dui in libero euismod commodo nec sit amet est. Etiam id ipsum aliquam, vehicula erat sit amet, consequat tortor.

Etiam facilisis lacus nec pretium lobortis. Praesent dapibus justo non efficitur efficitur. Nullam viverra justo arcu, eget egestas tortor pretium id. Sed imperdiet mattis eleifend. Vivamus suscipit et neque imperdiet venenatis. In malesuada sed urna eget vehicula. Donec fermentum tortor sit amet nisl elementum fringilla. Pellentesque dapibus suscipit faucibus. Nullam malesuada sed urna quis rutrum. Donec facilisis lorem id maximus mattis. Vestibulum quis elit magna. Vestibulum accumsan blandit consequat. Phasellus quis posuere quam.

> “Everybody should learn to program a computer, because it teaches you how to think.”

Vestibulum ullamcorper risus auctor eleifend consequat. Vivamus mollis in tellus ac ullamcorper. Vestibulum sit amet bibendum ipsum, vitae rutrum ex. Nullam cursus, urna et dapibus aliquam, urna leo euismod metus, eu luctus justo mi eget mauris. Proin felis leo, volutpat et purus in, lacinia luctus eros. Pellentesque lobortis massa scelerisque lorem ullamcorper, sit amet elementum nulla scelerisque. In volutpat efficitur nulla, aliquam ornare lectus ultricies ac. Mauris sagittis ornare dictum. Nulla vel felis ut purus fermentum pretium. Sed id lectus ac diam aliquet venenatis. Etiam ac auctor enim. Nunc velit mauris, viverra vel orci ut, egestas rhoncus diam. Morbi scelerisque nibh tellus, vel varius urna malesuada sed. Etiam ultricies sem consequat, posuere urna non, maximus ex. Mauris gravida diam sed augue condimentum pulvinar vel ac dui. Integer vel convallis justo.

Nam rutrum magna sed pellentesque lobortis. Etiam quam mauris, iaculis eget ex ac, rutrum scelerisque nisl. Cras finibus dictum ex sed tincidunt. Morbi facilisis neque porta, blandit mauris quis, pharetra odio. Aliquam dictum quam quis elit auctor, at vestibulum ex pulvinar. Quisque lobortis a lectus quis faucibus. Nulla vitae pellentesque nibh, et fringilla erat. Praesent placerat ac est at tincidunt. Praesent ultricies a ex at ultrices. Etiam sed tincidunt elit. Nulla sagittis neque neque, ultrices dignissim sapien pellentesque faucibus. Donec tempor orci sed consectetur dictum. Ut viverra ut enim ac semper. Integer lacinia sem in arcu tempor faucibus eget non urna. Praesent vel nunc eu libero aliquet interdum non vitae elit. Maecenas pharetra ipsum dolor, et iaculis elit ornare ac.

Aenean scelerisque ullamcorper est aliquet blandit. Donec ac tellus enim. Vivamus quis leo mattis, varius arcu at, convallis diam. Donec ac leo at nunc viverra molestie ac viverra nisi. Proin interdum at turpis at varius. Nunc sit amet ex suscipit, convallis ligula eu, pretium turpis. Sed ultricies neque vel mi malesuada, et mollis risus lobortis. Sed condimentum venenatis mauris, id elementum dolor gravida ac. Sed sodales tempus neque, quis iaculis arcu tincidunt ut. Donec vitae faucibus dui. In hac habitasse platea dictumst. Donec erat ex, ullamcorper a massa a, porttitor porta ligula.


SurveyForge

SurveyForge transforms survey creation through AI-powered intelligence, utilizing the Gemini API to generate high-quality survey questions from simple research prompts.                                                                               The application combines a React frontend with Node.js backend and MongoDB for flexible data storage, enabling real-time analytics and comprehensive insight generation. 

Projects

SaveDollar

SurveyForge

SurveyForge

Let’s talk... 💬