๐ About the Project
Smart Tutor was inspired by the growing need for accessible, intelligent, and real-time educational support โ especially in regions where personalized tutoring is limited. We envisioned a system where AI agents can step in to assist learners anytime, anywhere, through decentralized and secure communication.
๐ What We Built
Smart Tutor is a decentralized AI-powered tutoring system built using the Fetch.ai uAgents framework. It connects two intelligent agents:
- Student Agent: Sends academic questions.
- Tutor Agent: Receives questions and responds with answers fetched using Google's Gemini AI API.
This setup enables a real-time Q&A loop between the two agents, creating a seamless tutoring experience.
๐ง What We Learned
- How to build and run autonomous agents using the
uAgentsframework. - How to integrate external AI APIs like Gemini into agent logic.
- Real-world implementation of decentralized communication between AI agents.
- The importance of clear protocol design for agent interactions.
๐ ๏ธ How We Built It
- Framework: Fetch.ai's
uAgentsfor agent creation and communication. - AI Integration: Google Gemini API to generate responses for tutor queries.
- Environment Setup: Managed dependencies with Poetry and environment variables using
dotenv. - Communication Logic: Designed message schemas and handlers for interaction between student and tutor agents.
๐ง Challenges We Faced
- Resolving complex import path issues due to nested directory structures.
- Debugging environment setup and virtual environment inconsistencies.
- Handling agent address exchanges for real-time communication.
- Ensuring the Tutor Agent properly handles and responds to AI queries without timeouts.
๐ฎ Whatโs Next for Smart Tutor
Smart Tutor is just getting started! Hereโs what we plan to explore next:
- ๐ฑ Mobile Interface: Build a user-friendly mobile app for students to chat with their AI tutor anytime, anywhere.
- ๐ Subject Specialization: Add support for multiple tutor agents specialized in different subjects (Math, Science, Languages, etc.).
- ๐ง Smarter AI: Upgrade the tutor with memory and context awareness for better conversation flow and follow-up questions.
- ๐ Multilingual Support: Enable communication in multiple languages to make learning accessible globally.
- ๐ Learning Analytics: Track student questions and performance over time, giving insights into strengths and areas that need improvement.
- ๐ก๏ธ Privacy & Security: Implement end-to-end encryption and stronger data protection mechanisms.
- ๐งช Gamification & Quizzes: Add interactive quizzes and badges to make learning fun and engaging.
The goal is to create a smart, interactive, and scalable tutoring experience โ powered by autonomous agents and AI.
Built With
- gemini-api-(google-ai)
- poetry
- python
- python-dotenv
- uagents-framework
Log in or sign up for Devpost to join the conversation.