AI Research Mentor

An AI-powered learning platform that generates personalized STEM roadmaps, visualizes learning paths, and helps students explore complex topics through interactive knowledge graphs.

Comment

Inspiration

Learning a new STEM topic can be overwhelming. When I wanted to learn subjects like Computer Vision, Reinforcement Learning, and Signal Processing, I found that information was scattered across courses, blogs, papers, and videos. Beginners often don't know where to start or what prerequisites they need before moving to advanced concepts.

I wanted to build a tool that could act like a personal research mentor and automatically generate structured learning paths for any STEM topic.

What it does

AI Research Mentor generates personalized learning roadmaps using Google's Gemini API.

Users can enter any STEM topic, and the platform creates:

  • A structured learning roadmap
  • Topic dependencies and prerequisite relationships
  • Interactive visual knowledge graphs
  • Detailed explanations for each learning stage
  • Progress tracking for completed topics
  • Topic-specific subtopics and concepts

The roadmap is displayed as an interactive graph where learners can explore concepts step-by-step from fundamentals to advanced research areas.

How I built it

The project uses:

  • React + Next.js for the frontend
  • TypeScript for type safety
  • FastAPI for the backend
  • Google Gemini API for roadmap generation
  • React Flow for graph visualization
  • Tailwind CSS for UI design
  • Local storage for progress tracking

The backend communicates with Gemini to generate structured roadmap data, which is then transformed into an interactive graph on the frontend.

Challenges I ran into

One of the biggest challenges was generating meaningful and topic-specific roadmaps instead of generic learning paths.

Another challenge was creating an interactive graph experience that remains usable while displaying complex prerequisite relationships.

I also spent significant time debugging API integration issues, authentication problems, and ensuring the generated roadmap data could be converted into a clean graph structure.

Accomplishments that I'm proud of

  • Successfully integrated Gemini-powered roadmap generation
  • Built a responsive and interactive graph-based learning experience
  • Created a clean UI optimized for STEM education
  • Implemented dynamic topic exploration and progress tracking
  • Designed a system that can generate roadmaps for a wide variety of technical subjects

What I learned

Through this project I gained hands-on experience with:

  • Large Language Model integration
  • Prompt engineering
  • Graph visualization systems
  • FastAPI backend development
  • Full-stack TypeScript development
  • Building AI-powered educational tools

What's next

Future improvements include:

  • Real course and research paper recommendations
  • Personalized learning plans
  • Multi-agent roadmap generation
  • Study scheduling and reminders
  • Collaborative learning features
  • Adaptive roadmaps based on user progress

Built With

Share this project:

Updates