DEV Community: Mayank Kumar Gautam

SpendSense: Making Expense Tracking Simpler with Open-Source AI 💸

Mayank Kumar Gautam — Sun, 25 Oct 2026 11:08:00 +0000

This is a submission for the "Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady" (https://dev.to/challenges/hacktoberfest-week3-2026-10-19).

What I Built

Managing personal expenses can become surprisingly difficult. Small purchases add up, transaction histories get messy, and it's easy to lose track of where money goes each month.

That's why I came up with SpendSense, an AI-powered expense organizer designed to make tracking everyday spending simpler.

The idea is to help users understand their spending habits without spending too much time manually organizing every transaction.

SpendSense aims to help users:

💰 Organize expenses into categories such as food, travel, shopping, and bills.
🤖 Categorize transactions with AI to reduce repetitive manual work.
📊 Summarize spending patterns and show where money is going.
🔎 Identify spending trends across different categories.
💡 Generate practical insights that help users review their budgets and make informed decisions.

The goal isn't to tell people how they must spend their money. It's to make their spending easier to understand.

How I Built It

SpendSense is designed around a simple workflow: turn raw expense information into useful, understandable insights.

The planned workflow is:

Add expenses: Enter transaction details such as the description, amount, and date.
Categorize transactions: Use AI to suggest an appropriate category based on the transaction description.
Organize spending: Group transactions into categories to make the data easier to review.
Analyze expenses: Calculate totals and summarize spending across categories and time periods.
Generate insights: Present understandable summaries that help users recognize spending patterns.

For reliability, basic calculations such as totals and category-wise spending should use deterministic application logic rather than relying on an AI model to perform arithmetic.

Technology stack: [Add the actual programming language, open-weight AI model, framework, database, and libraries used.]

The initial version can focus on manual expense entry, AI-assisted categorization, and a simple spending dashboard. Additional features can be introduced gradually as the project develops.

Why Does Open Innovation Matter?

Personal finance tools should be adaptable, transparent, and accessible.

Open innovation can help make SpendSense more flexible by allowing developers to experiment with different AI models, improve transaction categorization, and contribute features that fit different needs.

Here's why that matters:

Experimentation: Developers can compare open-weight models and test different categorization approaches.
Transparency: Contributors can inspect how expenses are processed and how insights are generated.
Customization: Users and developers can adapt the application to different expense categories and workflows.
Community contributions: Developers can improve the interface, add import options, write tests, and fix bugs.
Deployment flexibility: Depending on the chosen model and infrastructure, AI processing may be possible locally or on a user's own server.

Open-source development also encourages people to learn from one another and build on existing work rather than starting from scratch.

For a personal finance project, privacy deserves particular attention. SpendSense should minimize the financial information it collects, avoid unnecessary data sharing, and clearly explain how information is processed.

My Agent Session

[Optional: Add your DevRelay agent session link or embed your saved agent session here.]

This section can document how an AI coding agent helped develop or improve SpendSense, if an actual session is available.

The Lazy Payoff 😴

The real benefit of SpendSense is reducing the effort required to keep expenses organized.

Instead of manually deciding which category every transaction belongs to, users could review AI-generated suggestions and correct them when necessary.

For example, if a user records 30 transactions per month and categorizing each transaction manually takes 15 seconds, that adds up to 7.5 minutes of categorization work.

If AI-assisted suggestions reduce the average review time, users could save some of that effort. The actual savings will depend on categorization accuracy and the amount of editing required.

This is an illustrative calculation, not a measured result from SpendSense.

The larger goal is to make expense tracking simple enough that people are more likely to maintain the habit consistently.

Prize Categories

[Add the applicable partner prize categories listed on the official Week 3 challenge page. Remove this section if you are not entering any partner categories.]

Final Thoughts

SpendSense is built around a straightforward idea: understanding your spending shouldn't require complicated spreadsheets or hours of manual work.

By combining practical expense tracking with AI-assisted categorization, I want to explore how open-source AI can make everyday financial tasks easier.

I'm aiming to keep the first version focused, useful, and easy to improve over time.

Because sometimes, the best automation isn't the one that does everything. It's the one that makes a small, repetitive task noticeably easier.

Thanks for reading! 🚀

devchallenge #hf26challenge #OpenSource #AI #PersonalFinance #Hacktoberfest

RepoPilot: An Open-Source AI Assistant for GitHub Issues

Mayank Kumar Gautam — Sun, 18 Oct 2026 14:10:05 +0000

This is a submission for the Hacktoberfest: Open-Source AI Challenge Week 2, hosted by DEV and its community partners.

What I Built

RepoPilot is an AI-powered assistant designed to help open-source maintainers manage GitHub issues more efficiently.

Open-source projects often receive bug reports, feature requests, and questions that take time to review and categorize. RepoPilot aims to simplify this workflow by analyzing issue descriptions, identifying the main problem, suggesting relevant labels, and generating helpful response drafts.

The goal is to make open-source contributions easier for maintainers and newcomers alike.

Demo

The demo will showcase how a user submits an issue description and receives an AI-generated summary, suggested category, priority, and response draft.

Code

The project repository will include the source code, installation instructions, environment configuration, and examples so that other developers can run and contribute to RepoPilot.

How I Built It

RepoPilot is designed to use an open-source language model through Hugging Face Transformers or a compatible inference provider.

The application can use Python for the backend, FastAPI for API endpoints, and a lightweight web interface for submitting issues and displaying results.

The workflow is straightforward:

The user submits a GitHub issue description.
The backend prepares the text for analysis.
An AI model summarizes the issue and suggests a category and priority.
The application generates a draft response for the maintainer.
The user reviews the suggestions before applying them.

The project is designed to be modular, allowing developers to replace the model, customize prompts, and add GitHub API integration.

Why Does Open Innovation Matter?

Open innovation allows developers to learn from one another, improve existing tools, and build technology that more people can access.

An open-source AI assistant gives contributors the opportunity to inspect the implementation, experiment with different models, improve prompts, and adapt the tool to their own projects.

Using open-source AI encourages transparency and experimentation instead of locking the entire workflow behind a proprietary API.

RepoPilot aims to demonstrate how accessible AI tools can help reduce repetitive work and make open-source collaboration more welcoming.

Thanks for supporting open-source AI and the developer community!!

Outside Quest 🌿 — An Open-Weight AI That Gets You Off Your Screen

Mayank Kumar Gautam — Sun, 11 Oct 2026 15:34:09 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass

What I Built

Outside Quest is an AI-powered outdoor adventure companion that helps people spend less time scrolling and more time exploring the real world.

We spend so much time looking at screens that sometimes we need a little push to step outside. Outside Quest turns that push into a small, personalized mission.

Users choose their surroundings, how much time they have, their energy level, and what interests them. The app then generates an outdoor quest they can complete in real life.

Some examples include:

🐦 Birdwatching: Obs