From README.md in the repository.
A Streamlit prototype that explains study topics through analogies drawn
from a learner's interests (Minecraft, Marvel, Cricket, Cooking, K-Pop),
then quizzes them in the same context. Built by Team Innovsphere during a
3-hour team sprint.
💡 The Problem
New ideas are easier to remember when they are tied to something you already
know well. Standard textbooks explain every topic the same way for every
reader.
🛠️ The Solution
PersonaLearn sends each topic to Google Gemini with a prompt that asks for
an explanation framed around the learner's chosen interest.
- Don't understand Trigonometry? Read an explanation based on Cricket.
- Confused by Blockchain? Learn it through Minecraft block analogies.
✨ Key Features
- Preset syllabi: Topic lists for School, SSC, UPSC, and General Knowledge.
- Interest mapping: Choose an analogy theme (Minecraft, Cricket, Marvel, Cooking, K-Pop) plus length and complexity.
- Two views:
- 🎓 Student view: Learn topics, mark them complete, and see a quiz score.
- 👨🏫 Professor view: Add or remove syllabus topics.
- AI quiz generator: Generates a multiple-choice question from the explanation that was just shown.
- Scoring: Running score and quiz count for the current session.
🚀 Quick Start
Prerequisites
Installation
Clone the repository
git clone https://github.com/Soham-o/PersonaLearn.git
cd PersonaLearn
Install dependencies
pip install -r requirements.txt
Set up credentials
Create a .env file in the root directory (do not commit it):
GOOGLE_API_KEY=your_api_key_here
If no key is set, the sidebar asks for one when the app starts.
Run the app
streamlit run main.py
📸 Screenshots
🏗️ How it Works
- Input: The user picks a topic (e.g., "Photosynthesis") and an interest (e.g., "Marvel").
- Processing: The app builds a prompt and sends it to Gemini through the
google-generativeai SDK (gemini-2.5-flash).
- Output: The interest-based explanation is displayed.
- Assessment: Gemini generates a quiz question based on that explanation.
⚠️ Limitations
- Prototype: built in a short sprint; there are no automated tests.
- No accounts or persistence: Student/Professor is a view toggle, not a login, and progress, scores and syllabus edits are lost when the session ends.
- Deprecated SDK:
google-generativeai no longer receives updates. Migrating to Google's google-genai package is recommended.
- AI output is not fact-checked: explanations and quiz answers come straight from the model.
👥 Team Innovsphere
Built during a 3-hour rapid development sprint by:
- Chaitanya
- Nishant
- Lavisha
- Ayush
- Soham
📄 License
No license file has been added yet. Until one is, all rights are reserved by
the authors.
