concept_bridge

ConceptBridge is an AI-powered GATE EE mentor that combines Gemini, textbook-grounded learning, and MongoDB memory to help students truly understand engineering concepts.

Comment

🌉 ConceptBridge AI Mentor

Inspiration

As a second-year Electrical Engineering student preparing for GATE, I constantly struggled with understanding how concepts connect. Most resources taught formulas, but not the reasoning behind them. AI tools could answer questions, but they often forgot previous conversations and didn't adapt to my learning progress.

I built ConceptBridge to solve a problem I personally face every day: learning Electrical Engineering concepts through intuition, connections, and long-term guidance instead of memorization.

What it does

ConceptBridge is an AI-powered GATE EE mentor built with Gemini and MongoDB.

Unlike a traditional chatbot, it:

  • Explains concepts using intuition first
  • Searches trusted EE textbooks using RAG
  • Identifies prerequisite knowledge gaps
  • Remembers student progress across sessions
  • Tracks weak topics and learning history
  • Creates a personalized learning experience

The goal is not just to answer questions, but to help students truly understand concepts.

How I built it

I used:

  • Gemini for reasoning and teaching
  • MongoDB Atlas for persistent student memory
  • ChromaDB for textbook retrieval (RAG)
  • FastAPI for the backend
  • HTML/CSS/JavaScript for the frontend

The system combines concept dependency tracking, textbook search, and long-term memory to create an AI mentor experience.

Challenges I ran into

The biggest challenge was reducing hallucinations in technical subjects. To solve this, I added textbook-based retrieval so explanations are grounded in trusted EE references.

Another challenge was designing a memory system that tracks learning progress instead of simply storing chat history.

What I learned

This project taught me the difference between a chatbot and an AI agent. A chatbot answers questions, while an agent remembers, adapts, and guides users toward a goal.

I also learned how to integrate Gemini, MongoDB, RAG pipelines, and FastAPI into a complete application.

What's next

I plan to expand ConceptBridge with:

  • More GATE EE subjects
  • Adaptive study plans
  • Automated revision scheduling
  • Mock tests and performance analytics

ConceptBridge started as a tool for my own GATE preparation, and I hope it can eventually help many other Electrical Engineering students learn more effectively.

Built With

Share this project:

Updates