👨‍🏫 AI Teaching Agent Team

A smart multi-agent system that creates personalized learning programs, roadmaps, curated resources, and hands-on exercises using AI + Google Docs — in one click.

---

🚀 Overview

AI Teaching Agent Team is an AI-powered educational tool that brings together four intelligent agents — each with a unique academic role — to create an end-to-end, customized learning experience on any topic you enter.

From building a foundational knowledge base, to designing a full roadmap, to curating real resources and crafting practice exercises — all results are neatly written to auto-generated Google Docs, so you can study or share them instantly.

---

🧠 Meet the AI Agents

Agent	Role	Responsibilities
🧑‍🏫 Professor	Knowledge Specialist	Builds a comprehensive report explaining the topic from first principles
📚 Academic Advisor	Roadmap Architect	Breaks the topic into a logical sequence of subtopics with time estimates
🔍 Research Librarian	Resource Curator	Finds high-quality blogs, videos, repos, docs, and links via SerpAPI
🧪 Teaching Assistant	Practice Designer	Crafts quizzes, exercises, and projects with solutions

All agents are powered by GPT-4o-mini and use tools like Google Docs, SerpAPI, and Composio.

---

✨ Features

• 📖 Auto-generates Google Docs for each learning section
• 🤖 Four collaborative AI agents, each with a specialized purpose
• 🔗 Uses Composio to integrate Google Docs programmatically
• 🔍 Leverages SerpAPI for real-time resource discovery
• ⚡️ Instant deployment via Streamlit UI
• 🧩 Modular and customizable agent structure

---

📂 Project Structure

📦 AI_teaching_agent_team/
├── 📁 agents/        # 🤖 Definitions for each specialized AI agent
│   ├── professor.py          # 👨‍🏫 "Professor" agent: creates the knowledge base
│   ├── academic_advisor.py   # 🧭 "Academic Advisor" agent: builds learning roadmaps
│   ├── research_librarian.py #🔍 "Research Librarian" agent: curates resources
│   └── teaching_assistant.py # 📝 "Teaching Assistant" agent: creates exercises & quizzes
│
├── 📁 tools/         # 🛠️ Tools used by agents
│   └── composio_tools.py # 🧰 Wrappers and helpers for Composio Google Docs API
│
├── 📁 config/
│   └── page_config.py # 📄 Streamlit page configuration settings
│ 
├── 📁 core/          # 🔧 Core app logic and shared session
│   ├── session_state.py # 🗃️ Handles Streamlit session state management
│   ├── api_validation.py # ✅ Validates and manages API key inputs
│   └── topic_input.py # 📝 Manages user topic input logic
│
├── 📁 ui/
│   ├── sidebar.py # 🖥️ Sidebar UI for API key inputs and settings
│   └── display_outputs.py # 🖼️ Logic to display agent outputs in the main app
│
├── 📁 utils/
│   └── doc_link_extractor.py # 🔗 Utility to extract Google Doc links from agent
│
├── 📄 app.py             # 🚀 Main Streamlit app – UI + Agent Orchestration
├── 📄 requirements.txt   # 📜 dependencies
├── 📄 LICENSE            # 📝 MIT License.
└── 📄 README.md          # 📘 Project documentation

---

🛠 Tech Stack

• Streamlit - Interactive UI
• Agno AI - Agent architecture
• OpenAI - GPT-4o-mini model
• Composio - Google Docs integration
• SerpAPI - Web search and resource discovery

---

🧪 Demo Preview

🔍 Topic: "LoRA for LLM fine-tuning"

📄 Professor: Google Doc with technical explanation
📄 Academic Advisor: Google Doc with a learning roadmap
📄 Research Librarian: Curated resources (YouTube, GitHub, blogs)
📄 Teaching Assistant: Exercises and project tasks

✅ All neatly structured. All in shareable Docs.

🧑‍💻 How to Use

1. Clone the Repo

git clone https://github.com/andredisa/AI_teaching_agent_team.git
cd ai-teaching-agent-team

2. Install Dependencies

pip install -r requirements.txt

3. Run the App

streamlit run app.py

4. Enter API Keys

• 🔐 OpenAI API Key

• 🔐 Composio API Key

• 🔐 SerpAPI Key

Input them securely from the sidebar UI.

5. Enter a Topic Type a topic like:

e.g., Large Language Models, Graph Neural Networks, Quantum Computing

Then click Start — and watch your personalized learning suite get built by the AI agents 🚀

---

💡 Future Ideas

• 🧠 Agent memory and personalization
  Let agents remember user preferences, learning pace, and previous topics for a more tailored experience.

• 🌍 Multi-language learning materials
  Automatically translate content and create learning docs in multiple languages to reach global learners.

• 🔄 Export to PDF and Notion
  Add functionality to export all generated content into formatted PDFs or directly publish into Notion pages.

• 💬 Conversational feedback from each agent
  Allow users to ask follow-up questions to each agent and receive interactive responses in a chat-like format.

---

✨ Contributing

🎉 Contributions are more than welcome!

If you find a bug 🐞, have a feature request ✨, or want to improve the code 💻:

• Open an Issue
• Submit a Pull Request 🚀

💬 Feel free to reach out on GitHub or by email!

Let’s build this together!

---

📜 License

📄 This project is released under the MIT License.
Please refer to the LICENSE file for full details.

---

☕ Support Me

If you find my work useful and would like to support me, you can buy me a coffee! Your support helps me keep creating and improving my projects. Thank you! 😊

---

🧑‍💻✨ Happy coding