Agentic AI describes systems that don't just answer a single prompt, but work toward a goal: they decide what to do next, use tools to act, look at the results, and keep going until the job is done. This module introduces the core idea and the vocabulary you'll use for the rest of the course — the difference between a chatbot, a fixed workflow, and a true agent, and the loop that drives every agent.
The first two lessons are short, high-level explanations of what agents are and how they're put together: a language model combined with tools, memory, and reasoning. The last lesson is a talk on agentic workflows — the iterative patterns (reflection, tool use, planning, and multi-agent collaboration) that show up again in later modules.
Watch the lessons in order. Don't worry if some terms are new — tool use, planning, memory, and multi-agent systems each get their own module later on.
Jeff Su's non-technical walkthrough of the three levels — plain LLMs, workflows, and agents — and what changes when an LLM, rather than a human, makes the decisions.
Video 10minIBM Technology's Maya Murad traces the shift from single models to compound AI systems, and shows how agents combine an LLM with tools, databases, and reasoning to solve harder problems.
Video 12minAndrew Ng's Sequoia AI Ascent talk on why iterating — drafting, reviewing, and revising — can beat a single prompt, and the design patterns behind agentic workflows.
Video 13min