🚀 Agentic AI Course: Building Intelligent Agents

Welcome to the Agentic AI Course! This repository is your hands-on laboratory for building and deploying Agentic AI. You will learn to orchestrate logic using Langflow and create interactive front-ends using Gradio.

---

📚 Course Overview

This course is structured into 5 sections. The first section covers environment setup, while the remaining 4 sections focus on building the following agents:

1. Agent 1: Exam Strategy Agent – An AI tutor that analyzes uploaded exams, provides feedback on student answers, and explains complex concepts.
2. Agent 2: YouTube Video Summary Generator – Converts long YouTube videos into short, actionable summaries to save study time.
3. Agent 3: Smart Homework Planner – Predicts the best study blocks by reading your calendar and assignment deadlines.
4. Agent 4: Personal Portfolio Website Agent – A developer agent that crafts a unique portfolio website based on your skills and interests.

---

📋 Pre-requisites

• MySphere Account: Ensure you are logged in at mysphere.net.
• GitHub Account: To fork this repo and run the environment.
• Google Gemini API Key: Required for the AI components within Langflow.

---

🛠️ Phase 1: Initial Environment Setup

Perform these steps once to get started.

1. Fork the Repository: Click the Fork button at the top right of this page.
2. Launch GitHub Codespaces:
   • Go to github.com/codespaces.
   • Select your fork and choose the 2-core CPU / 8GB RAM environment.
   • Click Create codespace and wait for the editor to load.

---

🤖 Phase 2: Running the AI Infrastructure (Langflow)

This starts the backend engine that manages your Agentic Flows.

1. Start Langflow: Run the following command in the terminal:

   docker compose up -d

2. Make Langflow Public:
   • Open the Ports tab in the terminal area.
   • Find Port 7860.
   • Right-click the "Visibility" column and select Make Public.
   • VS Code may notify you of a running application; you can ignore or close this notification.

---

🖱️ Phase 3: Running an Agent Example

You will repeat these steps for each of the 4 agents in the course.

1. Navigate to the Agent Folder:

   cd [folder-name]
   # Example: cd exam-strategy-agent

2. Create Python Environment, Activate Python Environment & Install Packages:

   python3 -m venv my_env
   source my_env/bin/activate
   pip install -r requirements.txt

3. Launch the Gradio UI:

   python3 app.py or python3 app_chat.py

4. Get the Public URLs:
   • Langflow URL: Copy the Public URL for Port 7860 from the Ports tab.
   • Gradio URL: Gradio has share=true enabled, so it will generate a https://...gradio.live link in the terminal. Use this .live link for registration.

---

🔗 Phase 4: Connecting & Registering

To link your local code to the MySphere course platform:

1. Configure IDs: Open the Langflow UI (Port 7860) and retrieve your Flow ID, Langflow API Key, and Google Generative AI Component ID.
2. Register the Codespace: Click the link below to open the registration form: 👉 Register your Codespace (Course ID: 442)
3. Submit URLs: Paste your Public Gradio URL and Public Langflow URL into the form and click Register.
4. Success: Once registered, you can interact with your agent!

---

⌨️ Terminal Command Reference

Command	Explanation
docker compose up -d	Runs the Langflow container in the background (detached mode).
pip install -r requirements.txt	Downloads and installs all Python libraries needed for the agent.
python3 app.py	Executes the Python script to launch the Gradio User Interface.
source env_name/bin/activate	Switches the terminal to use the specific Python environment for that folder.

Important

Switching Agents: To move to a different example, stop the current process (Ctrl + C), use cd .. to return to the root directory, navigate to the next agent folder, and repeat Phase 3 & 4.