StudyMate

Frontend of model

Inspiration

Students often struggle to decide what to study, how to divide their available time, and which tasks to complete first. We wanted to create a simple AI study companion that students could interact with naturally instead of using a complicated planning application.

Inspired by the conversational assistant experience of Alexa+, we built StudyMate, a web-based simulated Alexa+ experience that helps students plan and manage their studies through natural-language requests.

What it does

StudyMate is an AI-powered study companion that allows students to interact with it using simple natural-language requests.

It can:

Create personalized study plans based on subject, number of days, and available study hours. Add study tasks. Show current tasks. Mark tasks as completed. Answer study-related questions in simple language.

For example, a student can ask:

"Create a 5-day DBMS study plan."

or:

"Add Practice SQL to my tasks."

StudyMate understands the request and performs the appropriate action.

How we built it

We built StudyMate as a web application using Python, Flask, Google Gemini, HTML, CSS, and JavaScript.

The frontend provides the conversational interface. User requests are sent to the Flask backend, where Gemini analyzes the request and determines the appropriate action.

The backend then executes Python functions for actions such as creating study plans, adding tasks, viewing tasks, and completing tasks.

The basic workflow is:

User Request ↓ Web Interface ↓ Flask Backend ↓ Gemini Intent Understanding ↓ Python Action ↓ Result Displayed to User

The project is implemented as a simulated Alexa+ experience through our own web interface.

Challenges we ran into

One of the main challenges was connecting generative AI responses with actual application actions.

AI responses are not always returned in exactly the format an application expects. We solved this by designing structured prompts and using JSON responses so that the backend could identify the requested action and execute the appropriate Python function.

We also faced API availability and model-related issues during development and had to handle these errors without breaking the application.

Another challenge was building a useful prototype within the limited hackathon timeframe while keeping the application simple and reliable.

Accomplishments that we're proud of

We are proud of building a working AI study companion from scratch and connecting generative AI with real application actions.

StudyMate successfully demonstrates:

Natural-language interaction AI-based intent understanding Personalized study-plan generation Task management AI-powered study explanations Frontend and backend integration

We are especially proud of implementing an Understand → Decide → Execute → Respond workflow rather than creating only a static study-planning application.

What we learned

Through this project, we learned how to integrate generative AI into a real web application and connect AI-generated intent with backend functionality.

We gained practical experience with:

Gemini API integration Flask REST APIs Prompt engineering Structured AI responses Python backend development Frontend-backend communication Git and GitHub Conversational AI application design

The project also taught us that an AI application is more useful when the AI can understand a user's intention and connect it to an actual action.

What's next for StudyMate

We want to evolve StudyMate into a more complete personal study companion.

Future improvements include:

Persistent database storage User accounts and personalized profiles Study-progress tracking Reminders and notifications More intelligent study-plan generation Voice-based interaction Integration with learning resources More advanced agentic capabilities

Our current version focuses on demonstrating the core conversational AI experience, while these improvements could make StudyMate a more powerful long-term learning companion.