Build Your Own AI-Powered Learning Assistant with Claude & ChatGPT

Listen to this Post

Featured Image

Introduction:

With AI transforming education, why wait for schools to adopt advanced tools like Claude’s learning mode? By leveraging open-source platforms like Open WebUI, cloud hosting via Hostinger, and structured prompts, you can create a personalized AI tutor for students—with full control over security and functionality.

Learning Objectives:

  • Deploy Open WebUI on a VPS for a self-hosted AI learning assistant.
  • Configure custom system prompts for structured tutoring.
  • Restrict access to educational models only for safety.
  • Switch between Claude, ChatGPT, or other LLMs seamlessly.

1. Setting Up Open WebUI on a VPS

Open WebUI provides a ChatGPT-like interface but with self-hosted control. Here’s how to install it on a Linux VPS:

Installation Commands:

 Install Docker (if not already installed)
sudo apt update && sudo apt install docker.io -y

Run Open WebUI in a Docker container 
docker run -d -p 3000:3000 --name open-webui ghcr.io/open-webui/open-webui:main 

What This Does:

  • Installs Docker (required for containerized deployment).
  • Deploys Open WebUI on port 3000, accessible via http://<your-server-ip>:3000.

Next Steps:

  • Access the WebUI, create an admin account, and configure models.
    1. Hosting on a Cloud VPS (Hostinger Recommended)
      For optimal performance, use a cloud VPS like Hostinger:

Recommended VPS Setup:

  • OS: Ubuntu 22.04 LTS
  • Resources: At least 2 vCPUs, 4GB RAM (for smooth LLM inference).

Why Hostinger?

  • Affordable, 1-click VPS deployment.
  • Supports Docker and AI workloads efficiently.

3. Configuring Custom System Prompts for Tutoring

To replicate Claude’s learning mode, use structured prompts like Ethan Mollick’s “Tutor Prompt”:

Adding a System

  1. In Open WebUI, go to: Settings → Models → Claude/ChatGPT.

2. Paste the prompt (example below):

"You are an AI tutor. Start by assessing the student’s knowledge level. 
Ask probing questions, provide tailored examples, and never give direct answers. 
Encourage Socratic learning." 

Why This Works:

  • Forces AI to guide learning instead of spoon-feeding answers.
  • Mimics Claude’s educational approach.

4. Restricting Access to Educational Models Only

For child-safe AI, restrict access to only tutoring models:

Firewall Rule (Linux):

 Block non-educational API endpoints 
sudo iptables -A INPUT -p tcp --dport 3000 -m string --string "/api/chat" --algo bm -j DROP 

What This Does:

  • Prevents misuse by blocking non-tutoring API calls.

5. Switching Between Claude & ChatGPT

Open WebUI supports multiple LLMs. To switch:

Model Configuration:

  1. In Settings → Models, add API keys for:

– Claude (Anthropic API)
– ChatGPT (OpenAI API)

2. Select the preferred model per user.

Pro Tip:

  • Use Llama 3 (open-weight model) for cost-effective self-hosting.

What Undercode Say:

  • Key Takeaway 1: Self-hosting AI tutors enhances privacy & control over commercial solutions.
  • Key Takeaway 2: Structured prompts optimize learning by preventing over-reliance on AI answers.

Analysis:

The rise of self-hosted AI education tools signals a shift toward personalized, secure learning. Schools lag in AI adoption, but tech-savvy parents and educators can now deploy bespoke tutoring systems—balancing flexibility and safety.

Prediction:

By 2026, 50% of homeschooling setups will integrate self-hosted AI tutors, reducing dependency on traditional EdTech platforms. Expect more open-source tools for AI-driven education, with enhanced parental controls and local LLM optimizations.

Final Thought:

Why wait for the future? Build your AI tutor today—before schools catch up. 🚀

🎯Let’s Practice For Free:

IT/Security Reporter URL:

Reported By: Chuckkeith Why – Hackers Feeds
Extra Hub: Undercode MoN
Basic Verification: Pass ✅

🔐JOIN OUR CYBER WORLD [ CVE News • HackMonitor • UndercodeNews ]

💬 Whatsapp | 💬 Telegram

📢 Follow UndercodeTesting & Stay Tuned:

𝕏 formerly Twitter 🐦 | @ Threads | 🔗 Linkedin | 🦋BlueSky