Objectives:

To equip students to design, build and evaluate trustworthy applications powered by large language models. Participants learn Retrieval-Augmented Generation (RAG), agentic AI patterns, response-quality evaluation, hallucination reduction and responsible AI, then apply them hands-on to build a working software application in international teams.

Learning Outcomes:

Students can build a RAG pipeline, design agentic AI workflows, evaluate answer quality, mitigate hallucinations and apply responsible-AI principles in a functional software application.

Day 1: LLM and RAG foundations;

Day 2: retrieval and embeddings laboratories;

Day 3: agentic AI patterns;

Day 4: evaluation and hallucination reduction;

Day 5: team software application build and presentations.