-
-
Pipeline
-
Dashboard
Inspiration Students often struggle not because they lack learning resources, but because they lack proper planning, consistency, and organization. We wanted to build something that acts more like a real academic assistant instead of a normal chatbot. The idea behind StudyMate AI was to create an autonomous AI study agent that could help students manage their study workflow, generate personalized plans, track progress, and simplify productivity.
What it does StudyMate AI is an AI-powered study assistant that helps students create personalized study plans based on their goals and available time. It can:
Generate structured study roadmaps Recommend learning resources Track study progress Simulate integrations with Google Calendar and Gmail Store study plans and weak areas using a database Provide an interactive AI-powered study experience
The platform combines an intelligent frontend experience with a FastAPI backend powered by Gemini AI.
How we built it We built the project using a full-stack architecture: Frontend: HTML, CSS, JavaScript Backend: FastAPI (Python) AI Integration: Gemini API Database: SQLite Deployment: Vercel for frontend and Railway for backend Version Control: Git and GitHub
The backend was designed around an agent workflow system where the AI performs multiple tasks step-by-step such as parsing the user goal, searching resources, building study plans, tracking progress, and preparing integrations.
Challenges we ran into One of the biggest challenges was connecting the frontend and backend after deployment because both services were hosted separately. We also faced issues with: GitHub repository synchronization Railway deployment configuration
API response handling Dynamic frontend updates Managing real vs simulated integrations Proper UI rendering for AI-generated plans Another challenge was designing the project so it behaved like an autonomous AI agent rather than a simple chatbot.
Accomplishments that we're proud of We are proud that we successfully: Built and deployed a complete full-stack AI application Integrated Gemini AI into a working backend system Created a modern and responsive frontend UI Designed a multi-step AI workflow system Connected live frontend and backend deployments Added database-based study tracking functionality Created a project that feels like a real AI productivity tool
What we learned Through this project we learned: FastAPI backend development Frontend-backend communication using APIs Cloud deployment workflows Git and GitHub collaboration Railway and Vercel hosting AI workflow architecture API debugging and deployment troubleshooting Designing better user experiences for AI applications
We also learned how important system design and deployment architecture are when building real-world AI products.
What's next for Untitled Our future plans for StudyMate AI include: Real Google OAuth integration Live Google Calendar scheduling Gmail summary automation AI-generated quizzes and adaptive testing Progress analytics dashboards Personalized learning recommendations Mobile responsiveness improvements Multi-user authentication system We want to evolve StudyMate AI from a hackathon project into a fully functional AI-powered academic productivity platform.
Log in or sign up for Devpost to join the conversation.