"How Smart is AI? – Real projects to build real understanding"
summary
Shows a concrete classroom project: a Python Jupyter notebook plus template that calls free GitHub Models via an API and a personal access token, or runs locally with Ollama. Demonstrates building a pet‑shop RAG chatbot (keyword or vector search + concatenated sources), tweaks system prompts and temperature, and uses GitHub Copilot for code scaffolding. Claims GPT predicts text not facts, warns about hallucinated citations, and supplies runnable code and lesson-ready examples.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 1:09What is AI
- 8:11Generative AI basics
- 15:29Teaching AI projects
- 21:54Simple AI chatbot
- 25:53Chatbot loops
- 26:14RAG pattern
- 28:43Pet shop chatbot
- 30:30AI behavior
- 31:12Conclusion
related talks
12mUsing AI to Entice Employees to Python
45mBringing the Power of AI to Your Application
25mDemocratising learning of AI - experiences from the University of Oxford
10mMeet the Speakers
34mAI won't take your job but it sure will help you do it better
27mBuilding AI applications: Successes and Lessons
01h 01mUnlocking the Potential of AI
25mNo Code, No Problem: Building Powerful AI Agents with Just Your Words