DEV Community: Apoorva Bharekar

✨📚 From Long Notes to Quick Summaries: My AI Project 🤖🚀💡

Apoorva Bharekar — Fri, 23 Oct 2026 16:22:36 +0000

What I built

As a learner and aspiring developer, I wanted to make studying long notes a little easier. Reading and revising large amounts of material can take time, so I built an AI Study Notes Summarizer that turns study notes into concise summaries.

The app lets users paste their notes or upload TXT and PDF files, choose a short, medium, or long summary, and download the result.

Live Demo

🔗 Live app: [https://apoorvabha4710-tech-ai-study-notes-summarizer-app-iykc5p.streamlit.app/]

💻 GitHub repository: https://github.com/apoorvabha4710-tech/ai-study-notes-summarizer

How it works

I built the app with Python, Streamlit, and Hugging Face Transformers, using the sshleifer/distilbart-cnn-12-6 summarization model.

The app extracts text from uploaded files, splits longer notes into chunks, summarizes the chunks, and lets users download the generated summary.

Why open-source AI?

Open-weight models make AI more accessible to developers who want to experiment, learn, and build useful applications without relying entirely on a closed AI API.

This project runs inference using the model loaded in Python rather than sending each summarization request to a hosted inference API. That can help reduce per-request inference costs and gives developers more control over how the model is used.

The deployed demo runs on a hosted server, so uploaded notes are processed there. Running the app locally offers a different option for people who want to keep their notes on their own machine.

How it fits the Slow and Steady theme

The goal is simple: spend less time manually condensing long notes and more time understanding and revising the important ideas.

Instead of trying to do everything at once, I built the app step by step, tested its PDF and TXT uploads, and improved how it handles longer text.

What I learned

Building this project helped me practise Python, Streamlit, model integration, file processing, Git, GitHub, and deployment.

I also learned that working with AI involves more than calling a model. Handling input files, managing long text, testing errors, and documenting a project are all important parts of building a useful application.

What's next?

I'd like to improve the text-splitting logic, make summaries more coherent across chunks, and add automated tests.

Try it out

If you try the app, I'd love to hear what you think and what features would make it more useful for your study routine.

devchallenge #hf26challenge #hacktoberfest #python #ai