DEV Community: Priya Kumari

🦥 StudySloth AI — Your Open-Source AI Study Buddy

Priya Kumari — Sun, 25 Oct 2026 18:30:16 +0000

🦥 StudySloth AI: Slow Down, Learn Smarter with Open-Source AI

What if studying felt less overwhelming and revision became easier?

As students, we often have lots of notes but limited time to revise them. Reading long notes repeatedly can be tiring, and identifying weak topics is not always easy.

That inspired me to build StudySloth AI — an AI-powered study assistant designed to turn notes into useful revision material.

💡 What I Built

StudySloth AI helps students study smarter by transforming their notes into structured learning resources.

The goal is simple: spend less time organizing study material and more time understanding it.

✨ Key Features

📝 Quick Summary: Turn lengthy notes into concise summaries.
🗂️ Flashcards: Generate flashcards for active recall and revision.
🧠 Interactive Quiz: Test your understanding with AI-generated questions.
🎯 Weak Spot Finder: Identify topics that may need more attention based on quiz performance.
🚀 Sloth Study Sprint: Bring your revision activities together in one focused study workflow.
🔎 Source Check: Help students compare generated answers with their original notes instead of blindly trusting AI output.

🛠️ How I Built It

I designed StudySloth AI around an open-source and local-first AI approach.

Planned technology stack:

Frontend: React and Vite
Backend: Python and FastAPI
AI model: Qwen2.5
Local inference: Ollama
Version control: Git and GitHub

The intended workflow is straightforward:

A student enters their notes.
The frontend sends the notes to the backend.
The backend communicates with the local AI model.
The generated summary, flashcards, or quiz is returned to the interface.
Students can use their quiz results to decide what to revise next.

Running inference locally can also reduce dependence on paid cloud AI APIs, although it requires a compatible local setup and sufficient computing resources.

🌱 Why Open-Source AI Matters

Open-source tools make it easier for developers and students to experiment, learn, and build useful applications without depending entirely on proprietary services.

By exploring open-weight models and local inference, I wanted to understand how AI-powered applications work beyond simply calling an API.

This project also gave me an opportunity to explore the connection between a frontend, a backend API, and a language model.

📚 What I Learned

Working on StudySloth AI helped me explore several important development concepts:

Connecting a React frontend to a backend API.
Structuring backend endpoints with FastAPI.
Integrating a language model into an application.
Designing practical AI features around a real student problem.
Thinking about reliability, structured responses, and grounding answers in source notes.
Organizing a project for GitHub and open-source collaboration.

One important lesson is that an AI application should not just generate impressive-looking answers. It should also help users verify those answers and understand their limitations.

🔗 Project Links

GitHub Repository: https://github.com/priyarani8281/dev3

Live Demo: https://drive.google.com/file/d/1NATNmo3SaseCFF15iIfstqTA2cOnH-Rn/view?usp=drivesdk

🚀 What's Next?

I would like to improve StudySloth AI by adding better quiz evaluation, more reliable source checking, improved error handling, and a smoother study experience.

I also want to continue learning how open-weight AI models can be used to build practical tools for students.

StudySloth AI is my step toward building AI applications that are useful, accessible, and designed around real problems.

Slow down. Learn smarter. One study sprint at a time. 🦥🌿

OpenSource #AI #StudySlothAI #Qwen #Ollama #React #FastAPI #StudentDeveloper

FileSense Local — Find Any Document Instantly

Priya Kumari — Sun, 18 Oct 2026 19:44:26 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle

What I Built

FileSense — Find Your Documents Without the Hassle

Have you ever needed an important PDF or assignment but couldn't remember where you saved it? I've faced this problem, and that's what inspired me to build FileSense.

FileSense is a document search tool that helps users find files on their computers without having to open folder after folder. I've built the search and indexing functionality, and my goal is to make finding documents quicker and easier.

I started this project because I wanted to build something that solves an actual everyday problem instead of creating another project just for my portfolio.

I can see students using it to find assignments and notes, freelancers searching for client files, and researchers managing their documents.

For the side-hustle part, I'd like to explore turning FileSense into a product with basic search available for free and more advanced features offered as a paid option in the future.

Demo

🎥 Video Demo: [https://drive.google.com/file/d/1PalvxZ9O5f4kxKFXHfPYl-OUw9HjzhL4/view?usp=drivesdk]

Code

💻 GitHub Repository: [https://github.com/priyarani8281/FileSense-Local]

How I Built It

I built FileSense with a focus on making document search simple and useful.

The current version includes document indexing and search functionality. I'm working on improving the experience so users can find their files more easily without manually checking every folder.

I'm also exploring how open-source AI models could make FileSense smarter by allowing users to search for documents based on their meaning, rather than only remembering exact filenames or keywords.

I want to keep the project practical and avoid adding AI just for the sake of calling it an AI project. Every feature should solve a real problem.

Why Does Open Innovation Matter?

For me, open innovation means having the freedom to learn, experiment, and build something useful using tools that other developers can also access and improve.

With open-weight AI models, I can experiment with different approaches, understand how they work, and potentially run AI features locally instead of depending entirely on paid APIs.

This is especially interesting for a document search tool because people may have personal or important files they don't want to upload to external services.

I'm still learning, and open-source tools give me the opportunity to explore these ideas without needing a big team or an expensive setup.

Building FileSense has made me think more about how small problems can lead to useful product ideas.

I'm still improving it, and I'd love to hear from people who deal with lots of files every day.

If you could add one feature to FileSense, what would it be? I'd love to know what would actually make it useful for you!

TouchGrass AI:Using Open-Source AI to spend Less Time on Screens

Priya Kumari — Sun, 11 Oct 2026 21:18:04 +0000

This is a submission for the "Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass" (https://dev.to/challenges/hacktoberfest-week1-2026-10-05)

What I Built

🌱 TouchGrass AI is an AI-powered web application designed to help students spend less time scrolling on screens and more time enjoying the outdoors.

As students, we often spend hours in front of laptops and phones. Sometimes, we want to take a break but don't know what to do with our free time. I built TouchGrass AI to make that decision easier!

Users can choose a break duration of 10, 20, or 30 minutes and receive AI-generated outdoor activity suggestions. They can start a timer, mark activities as completed, and track their progress.

Key Features

🌿 AI-generated outdoor activity suggestions
⏱️ Flexible 10, 20, and 30-minute breaks
✅ Activity completion tracking
📈 Progress tracking saved in the browser
🤖 Local AI inference using an open-weight model

The goal is simple: use AI to encourage people to step away from their screens and reconnect with the world around them.

Demo

Local demo: The app currently runs locally on my computer.

To try it, the project needs Node.js, Ollama, and the Qwen 2.5 3B model installed.

A publicly accessible live demo is not available yet.

Code

🔗 GitHub Repository: [https://github.com/priyarani8281/TouchGrassAI]

The repository contains the frontend, Express backend, and setup instructions.

How I Built It

I built TouchGrass AI using a combination of web technologies and open-source AI tools.

Tech stack:

HTML, CSS, and JavaScript for the frontend
Node.js and Express for the backend
Ollama for running AI locally
Qwen 2.5 3B, an open-weight language model, for generating outdoor activity suggestions

The frontend sends a request to the Express backend with the selected break duration. The backend communicates with the locally running Qwen model through Ollama and returns activity suggestions to the frontend.

I also added a timer, activity completion controls, and browser-based progress tracking.

As a beginner, building this project helped me understand how a frontend, backend API, and local AI model can work together in one application.

Why Does Open Innovation Matter?

Open innovation made it possible for me to experiment with AI without depending on a paid, proprietary AI API.

Using an open-weight model through Ollama allowed me to run inference locally, explore how AI requests work, and build an application while learning about the technology behind it.

It also makes the project easier for other developers to explore, learn from, and improve. I hope to continue developing TouchGrass AI and welcome feedback and contributions from the community.

For me, open innovation is about making technology more accessible and giving beginners the opportunity to learn by building real projects.

My Agent Session

I used OpenCode to help me build and improve the project.

I haven't published an agent session link yet.

Prize Categories

I am submitting TouchGrass AI for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass.

I'll select any additional partner prize categories only if the project meets their requirements.

Thanks for checking out TouchGrass AI! 🌱

I'd love to hear your suggestions: What would you add to help people spend more time outdoors and less time on their screens?

TouchGrass AI:using Open-Source AI to spend less time on screens

Priya Kumari — Sun, 11 Oct 2026 21:12:48 +0000

This is a submission for the "Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass" (https://dev.to/challenges/hacktoberfest-week1-2026-10-05)

What I Built

🌱 TouchGrass AI is an AI-powered web application designed to help students spend less time scrolling on screens and more time enjoying the outdoors.

As students, we often spend hours in front of laptops and phones. Sometimes, we want to take a break but don't know what to do with our free time. I built TouchGrass AI to make that decision easier!

Users can choose a break duration of 10, 20, or 30 minutes and receive AI-generated outdoor activity suggestions. They can start a timer, mark activities as completed, and track their progress.

Key Features

🌿 AI-generated outdoor activity suggestions
⏱️ Flexible 10, 20, and 30-minute breaks
✅ Activity completion tracking
📈 Progress tracking saved in the browser
🤖 Local AI inference using an open-weight model

The goal is simple: use AI to encourage people to step away from their screens and reconnect with the world around them.

Demo

Local demo: The app currently runs locally on my computer.

To try it, the project needs Node.js, Ollama, and the Qwen 2.5 3B model installed.

A publicly accessible live demo is not available yet.

Code

🔗 GitHub Repository: [https://github.com/priyarani8281/TouchGrassAI]

The repository contains the frontend, Express backend, and setup instructions.

How I Built It

I built TouchGrass AI using a combination of web technologies and open-source AI tools.

Tech stack:

HTML, CSS, and JavaScript for the frontend
Node.js and Express for the backend
Ollama for running AI locally
Qwen 2.5 3B, an open-weight language model, for generating outdoor activity suggestions

I also added a timer, activity completion controls, and browser-based progress tracking.

As a beginner, building this project helped me understand how a frontend, backend API, and local AI model can work together in one application.

Why Does Open Innovation Matter?

Open innovation made it possible for me to experiment with AI without depending on a paid, proprietary AI API.

Using an open-weight model through Ollama allowed me to run inference locally, explore how AI requests work, and build an application while learning about the technology behind it.

It also makes the project easier for other developers to explore, learn from, and improve. I hope to continue developing TouchGrass AI and welcome feedback and contributions from the community.

For me, open innovation is about making technology more accessible and giving beginners the opportunity to learn by building real projects.

My Agent Session

I used OpenCode to help me build and improve the project.

I haven't published an agent session link yet.

Prize Categories

I am submitting TouchGrass AI for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass.

I'll select any additional partner prize categories only if the project meets their requirements.

Thanks for checking out TouchGrass AI! 🌱

I'd love to hear your suggestions: What would you add to help people spend more time outdoors and less time on their screens?

DayTrack:Never miss an important reminder

Priya Kumari — Sun, 04 Oct 2026 21:04:12 +0000

Build for a Friend — Smart Reminder & Task Tracker

What I Built

I built a smart reminder and task-tracking web app for a friend who often has multiple things to do every day but sometimes forgets or misses them.

The idea is simple: instead of a reminder that only tells you "Do this task", the app also remembers what happened to the task.

For example, if I set a daily goal:

"Solve 3 DSA problems every day."

and I don't complete it today, the task is marked as missed. On the next day, the app reminds me about the missed task along with my new tasks.

The goal is to make the reminder system more useful for students who are trying to build consistent habits such as:

Solving DSA problems
Studying for exams
Learning a new technology
Completing assignments
Reading or practicing every day
Following personal goals

Instead of simply showing notifications, the app helps users understand what they planned, what they completed, and what they missed.

Demo

🎥 Video Demo: [https://drive.google.com/file/d/1kcYuEGgmPqv-AeIibYq4cts2bH7sjuiH/view?usp=drivesdk]

The demo shows the complete flow:

Register → Login → Create Task → Set Reminder → Complete/Miss Task → Track Progress → Carry Forward Missed Tasks

Code

💻 GitHub Repository: [https://github.com/priyarani8281/daytrack]

The project is built as a web application with a focus on making the interface simple and easy to use.

How I Built It

I built the project as a student-focused productivity application using modern web technologies.

The main parts of the application include:

Frontend: HTML, CSS, JavaScript
UI: Responsive and user-friendly interface
Task Management: Create, update, complete and track tasks
Reminder System: Reminders based on the user's selected time/date
Missed Task Tracking: Keeps track of tasks that were not completed
Carry-Forward Logic: Suggests previously missed tasks on the following day
Authentication: User registration and login
Dark/Light Mode: User-selectable interface theme
Persistent User Data: User information and tasks are stored so they are available when the user logs in again

For the AI-assisted development part, I used to help with development, debugging, understanding implementation ideas, and improving parts of the application.

The important part for me was not simply generating a project with AI. I used AI as a development partner while making the product decisions myself and testing the features as I built them.

Why Does Open Innovation Matter?

Open innovation makes it possible for individual developers and students to experiment with technologies that would otherwise require a lot of resources or expertise.

For this project, open-source AI and developer tools helped me move faster while still allowing me to understand and modify the underlying implementation.

A closed API can provide a useful service, but open models and open-source tools give developers more control over:

How the technology is used
How it can be customized
How it can be integrated into a project
How developers can learn from the technology
How the community can improve and extend it

For a student developer, this is especially valuable. Instead of only being a consumer of AI tools, I can experiment with them, understand how they work, and build something useful around them.

That is what I see as the real value of open innovation: making experimentation and building accessible to more people.

My Agent Session

I used the agent as a development partner during the project, particularly for debugging, implementation ideas, and improving the application.

Prize Categories

I am entering this project for:

Build for a Friend

Final Thoughts

This project started with a simple problem: reminders tell us what we should do, but they don't always help us deal with what we didn't do.

I wanted to build something that treats missed tasks as part of the user's progress instead of simply forgetting them.

Building this project also helped me learn more about web development, debugging, user authentication, persistent data, and using AI tools as part of a real development workflow.

It is still a work in progress, but that is also what makes the project exciting — there are many features I would like to add in the future, such as smarter task prioritization, streaks, analytics, recurring schedules, and personalized recommendations.

Thanks for checking out my project! ❤️