Deadline Decoder: Turn Confusing College Notices Into Clear Actions

This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend

What I Built

College students constantly receive notices about assignments, registrations, workshops, exams, and club activities.

The information is usually there, but figuring out what actually needs to be done, where to submit it, what is required, and when it is due can still be frustrating.

So I built Deadline Decoder — a tool that turns confusing college notices into clear actions.

You can:

• Paste a college notice
• Upload a PDF
• Upload a screenshot

Deadline Decoder extracts:

• The actual deadline
• What the student needs to do
• Submission method
• Required materials
• Important warnings
• Priority when explicitly mentioned

I built it for fellow college students and friends who regularly deal with confusing college notices.

The idea is simple:

Don't just read the notice. Decode it.

Demo

https://drive.google.com/file/d/1Duzi3bGLD_WuLX_u6rjADWa8xddXVeKm/view?usp=sharing

The demo shows Deadline Decoder taking a college notice and turning it into a structured result containing the deadline, task, submission method, required materials, and warnings.

Code

TanyaPandey25 / deadline-decoder

Turn confusing college notices into clear actions.

# Deadline Decoder

Turn confusing college notices into clear actions.

College notices are often full of dates, instructions, submission methods, requirements, and warnings. The problem is that students usually don't need a summary — they need to know exactly **what they have to do and by when**.

**Deadline Decoder** turns a college notice into a structured action-oriented result.

## What it does

You can provide a college notice in three ways:

- Paste the notice as text

- Upload a PDF

- Upload a screenshot/image

Deadline Decoder extracts:

- Deadline

- Tasks

- Submission method

- Required materials

- Important warnings

- Priority when explicitly mentioned

The goal is simple:

**Don't just read the notice. Decode it.**

## Why I built it

College students regularly receive notices for assignments, registrations, events, exams, workshops, and club activities.

The information is usually there, but it is not always easy to identify what…

View on GitHub

The project is open source and contains both the React frontend and FastAPI backend.

How I Built It

Deadline Decoder is built around Gemma, an open model from Google, accessed through the Gemini API.

The frontend is built with React and Vite.

The backend uses Python and FastAPI. PDF notices are processed using PyMuPDF, while text and image inputs are sent to Gemma for interpretation.

Gemma is responsible for the core intelligence: understanding the notice and extracting the information that actually matters to the student.

I also instructed the model not to invent information. If something such as a deadline isn't present in the notice, the result should say "Not specified" rather than guessing.

The project currently supports three input formats:
Text
PDF
PNG/JPEG screenshots

Why Does Open Innovation Matter?

For a project like Deadline Decoder, the AI model is not just an extra feature — it is the part that understands the messy, natural-language college notice and converts it into structured actions.

Using an open model such as Gemma makes it possible to build around an AI model that developers can access, experiment with, and integrate into their own applications rather than treating AI as a black-box chatbot experience.

It also makes the project more interesting from a development perspective: the application logic, prompts, structured output, and user experience can all be designed around the model.

For me, open innovation means being able to take an open model and turn it into something specific and useful for an everyday problem.