Alexa+ for students

Your Autonomous AI Copilot for Smarter Student Life.

Comment

Inspiration

Students often struggle to balance assignments, exams, deadlines, and personal responsibilities. Traditional task managers can remind students what to do, but they do not always help them decide what to do next.

We built Mentor AI to address this problem: an AI-powered educational copilot that helps students organize their academic responsibilities, plan their day, and make more informed decisions about their time.

What It Does

Mentor AI combines AI-powered assistance with structured task management and scheduling. Students can interact with the AI Copilot using natural language, manage academic tasks, generate study plans, and review proposed schedules before confirming them.

Our goal is to make academic planning more proactive, personalized, and manageable for students.

How We Built It

We built Mentor AI using Next.js and TypeScript for the frontend and FastAPI with Python for the backend. The system uses a database layer with SQLAlchemy and Alembic for data management and migrations.

For AI-powered assistance, we designed a provider-independent architecture supporting Google Gemini, Groq, and an offline rule-based fallback. We separated AI-generated planning suggestions from deterministic scheduling logic to make scheduling more predictable and easier to validate.

We also implemented structured task management, plan previews, explicit user confirmation, and staged actions to help keep important operations under user control.

Challenges We Faced

One of our biggest challenges was making AI assistance useful without allowing unpredictable AI outputs to control every operation. We addressed this by separating language understanding and plan generation from deterministic scheduling and validation.

We also worked on provider fallback behavior, API integration, authentication, data isolation, and reliable task state transitions. Balancing a polished user experience with backend reliability and security was another important challenge.

What We Learned

Building Mentor AI taught us that an effective AI product requires more than connecting a language model to a chat interface. It needs reliable system design, clear validation rules, thoughtful user experience, and appropriate human oversight.

We gained practical experience in full-stack development, AI provider integration, API design, database migrations, automated testing, and designing AI-assisted workflows.

What's Next

Our future roadmap includes an Academic Drift Detector that compares planned and actual progress and proposes recovery plans when students fall behind. We also plan to explore an Aura Score and Student Impact Score to recognize consistency and encourage productive study habits.

These are planned enhancements that we hope to develop in future iterations.

Our Vision

We envision Mentor AI as more than a task manager: a dependable educational copilot that helps students understand their workload, take the next best step, and build better academic habits—while keeping students in control of their decisions.

Built With

  • a
  • and
  • built-with-next.js
  • deterministic
  • fastapi
  • google-gemini-api
  • groq-api-(llama-3.3-70b)
  • postgresql/sqlite
  • python
  • scheduling
  • sqlalchemy
  • tailwind-css
  • typescript
Share this project:

Updates

Submission history