If you want to learn:
How do AI agents work and what makes them autonomous?
What is tool calling in LLMs and how does it actually function behind the scenes?
How can you chain multiple LLM calls to create more controlled AI workflows?
What is an agentic loop and how does it enable AI agents to execute complex tasks?
How do agentic workflows differ from traditional automation tools?
Then this lecture is for you!
This lecture breaks down the core mechanisms behind agentic AI systems and autonomous agents. You'll discover how LLM chaining works by splitting complex prompts into separate, controllable workflow steps that can be tested and optimized individually. The lecture demystifies tool calling by revealing the prompting techniques that allow AI agents to interact with external tools and APIs—showing you the exact input and output patterns that create this seemingly magical capability. You'll learn how agentic loops enable AI agents to autonomously execute multi-step tasks by repeatedly calling an LLM with updated context until a goal is achieved. Through practical examples like portfolio valuation and stock price lookup, you'll understand how agents work by combining tool invocation, decision-making, and iteration within a single workflow. The lecture provides hands-on demonstrations using ChatGPT to illustrate how tool use actually functions through clever prompt engineering rather than special LLM capabilities. By the end, you'll have a clear understanding of agentic workflows and how these autonomous AI systems coordinate multiple specialized agents to automate complex tasks without requiring human intervention at each step.