Programme Content
The course is structured to move participants from concept to implementation in a step-by-step manner. The first day builds the foundations of AI, LLMs and Agentic AI - how AI, machine learning, deep learning, NLP and generative AI relate to one another, how large language models actually process text, what tokens and context windows are, and why hallucinations happen. Participants then move into the LLM ecosystem, covering RAG (Retrieval-Augmented Generation), the difference between prompt engineering and loop engineering, and when a business use case needs a knowledge base rather than simple prompting. The day closes with the anatomy of an AI agent - brain, memory, tools and loop - and an introduction to vector databases. The second day focuses on frontier AI platforms and how they are used as working environments rather than chat windows: MCP (Model Context Protocol) and connectors, Claude as an agentic platform through Projects, Skills and agent building, Kimi, and the role of APIs in connecting AI to existing business systems. The third day introduces the no-code platform landscape through live demos - n8n/Make.com for automation and agent building, Lovable and Replit for website and interface generation through prompting, voice agent platforms such as Bolna, Vapi and Retell, and Landbot for chatbots - so that participants can judge which tool fits which business need. The fourth day is a full offline build day on constructing AI agents in n8n/Make.com, covering workflows, triggers and tools, building and testing working agents, and creating a voice agent on Bolna.ai that places real calls. The fifth day is the second offline build day, focused on multi-agent systems in n8n and on integration - participants connect a website, a chatbot and a voice agent into a single working solution, and finish with testing, error handling, safety checks and demo preparation.
