What if LLMs can do anything? - create team of agents Tool use and function calling Planning and reasoning systems ReAct prompting and chain-of-thought Hands-on: Create a task automation agent