Learning to Talk to Machines: Prompt Engineering and Reasoning with LLMs

Artificial Intelligence has moved from science fiction into everyday life, transforming how people learn, create, and interact. This Summer School explores AI’s evolution, from its early visions to today’s Generative AI systems that can write, draw, reason, and converse. Students will learn the basics of neural networks and large language models, how these systems learn from data, and how prompting can guide them effectively. Through demonstrations, experiments with public GenAI tools, and practical exercises, participants will examine both the promise and current limits of these technologies. The course concludes by looking ahead to agentic AI and future human-AI collaboration responsibly.

Through lectures and workshops, this Summer School offers an introduction to Artificial Intelligence with a focus on the recent breakthroughs in Generative AI and Large Language Models (LLMs). Beginning with a historical overview of AI and its key paradigms—symbolic reasoning, machine learning, and neural architectures—students will develop a conceptual understanding of how machines learn, represent knowledge, and generate creative outputs. The course places particular emphasis on the evolution of neural networks and the emergence of transformer-based architectures that underpin today’s generative systems.

The second part of activities emphasizes practical engagement and critical reflection. Students will experiment with state-of-the-art GenAI tools to understand how prompting strategies influence model behavior and reasoning. Topics such as Chain-of-Thought prompting, contextual reasoning, and multi-step task decomposition will be introduced, drawing on recent scientific literature. Through guided exercises, discussions, and demonstrations, participants will gain insight into both the capabilities and limitations of large-scale AI systems, and will conclude with a forward-looking exploration of agentic AI and autonomous reasoning frameworks.