AI-Assisted Programming
AI coding assistants are becoming a standard part of software development. But using them effectively requires more than typing prompts and accepting suggestions. Uncritical use can introduce subtle bugs, erode your own understanding of the code, and create long-term maintenance problems.
This course gives you the practical skills to work with AI tools productively and responsibly. We cover how to set up your environment, write prompts that actually work, manage context, and critically evaluate what AI produces. Equally important: we look at where AI assistance goes wrong, how it affects your learning, and what that means for the quality of your software and your own skills over time.
Sessions are interactive throughout: expect live polls, hands-on exercises in Python, and discussion of participants' own results.
This is a sequential course. Enrollment is limited to 50 participants.
Agenda
• Day 1 (24.08.): Before You Prompt: Setup, Safety & Responsible Use
- Chat interfaces vs. agentic tools: choosing the right setup
- Security: handling secrets and credentials when working with AI
- Copyright, licensing, and academic disclosure requirements
- Reproducibility and non-determinism in AI use
• Day 2 (31.08.): Talking to Machines: Prompts and Context Control
- Prompt engineering
- Context management and conversation structure
- Version control for AI-assisted workflows
• Day 3 (07.09.): Beyond the Hype: Technical Debt, Deskilling & Debugging
- Technical debt and knowledge debt
- Evaluating and debugging AI-generated code
- Deskilling: what the research says about AI and learning
Prerequisites
• Basic working knowledge of Python
• No prior experience with AI coding tools required
• Bring your own device with internet access and a working Python environment
• An account with an AI assistant of your choice, free tiers are sufficient
