• 

  application

• 

  Architecture of software

GradeGenius

From Paperwork to Personalization – AI for Teachers!

Inspiration

It all began on a typical college afternoon. We watched one of our favuorite teachers rushing between classes, arms full of assignments, her phone buzzing with student doubts, and her laptop screen cluttered with tabs for grading, lecture slides, and feedback forms. She smiled through it, but we could tell it was exhausting.

That’s when we started noticing a pattern. Many teachers around us were not just teaching; they were managing classrooms, creating study materials, handling doubts, grading hundreds of papers, and still expected to give each student personalized feedback. It wasn’t just overwhelming it was unsustainable.

We asked ourselves: What if we could help them?
What if there was a tool that could take care of repetitive tasks, so teachers could focus on what truly matters teaching and connecting with students?
That’s how GradeGenius was born a digital assistant that transforms paperwork into personalization.

What it does

GradeGenius is an AI-powered platform built to support overburdened teachers by automating assignment grading, enabling classroom management, and delivering personalized feedback. Here's how it works:

1. Create Kaksha:
   Teachers can create digital classrooms (Kaksha), invite students, post assignments, share notes, schedule online lectures, and resolve doubts. It creates an organized, interactive space for both teaching and learning. Students can also help each other by answering peers’ questions.

2. Generate PPT:
   With just a link or a document and a topic description, teachers can generate ready-to-use, customizable PowerPoint presentations using AI saving hours of preparation time.

3. Generate Quiz and Assignment:
   Teachers can auto-generate subject-specific quizzes and assignments. These are fully customizable, allowing teachers to tailor the difficulty, topic coverage, and format as needed.

4. Auto Grading with Personal Feedback:
   GradeGenius uses AI to assess submitted quizzes and assignments, assigns marks, and generates detailed, personalized feedback for each student. It also generates a performance report highlighting students who may need more attention, enabling early intervention.

How we built it

1. Analyzing the Root Cause:
   We spoke to multiple teachers to understand the pain points in their day-to-day academic life. The lack of time for personalized attention and repetitive administrative tasks emerged as major challenges.

2. Finding the Solution:
   After careful analysis, we mapped out a set of features that would significantly reduce the manual workload while increasing student engagement and performance tracking.

3. Tech Stack and Architecture:

   • Frontend: Flutter – for smooth, cross-platform mobile app development.
     
   • Backend: Google Cloud Platform (GCP), Firebase – for serverless infrastructure, user authentication, and real-time data sync.
     
   • Authentication: OAuth2 – for secure user access and identity management.
     
   • Database & AI/ML Layer: Google AI, Gemini, TensorFlow – to power intelligent feedback, grading, content generation, and predictive analytics.

Challenges we ran into

• Accurate Feedback Generation: Ensuring AI-generated feedback was personalized and contextually correct required extensive fine-tuning of the language model.
  
• Balancing Automation and Control: Giving teachers full control while still leveraging automation was tricky—we had to iterate UI/UX several times to get it right.
  
• Data Privacy: Since we are dealing with student data and performance, securing that data became a top priority. We integrated role-based access and encrypted sensitive data.

What we learned

• Empathy is a great design tool talking to real users (teachers) gave us insights we couldn’t have imagined on our own.
  
• Integrating AI isn’t just about automation; it's about making human processes better, not replacing them.
  
• Communication and collaboration were key we learned to work effectively as a team, balancing frontend, backend, and ML tasks seamlessly.

What's next for GradeGenius

• Automating the process for exam sheets grading using opencv technology, where the students will submit their sheets online and the AI will grade them accordingly.
• Creating a customizable dashboard for each kaksha, where the teacher can view the students progress in graphical way and also filter out results.

Built With

• bert
• firebase
• flutter
• gcp
• gemini
• mongodb
• tensorflow