Introduction to AI Agents

What You Will Learn

Upon completing the course, students will gain an in-depth understanding of how AI agents work and will be able to develop effective frameworks to create advanced agent AI systems across various domains and for tackling different tasks.

Course Requirements

• The course does not require prior programming knowledge or skills.
• If you are unfamiliar with prompt engineering techniques for LLM, it is highly recommended to take the courses "Introduction to Prompt Engineering" and "Advanced Prompt Engineering".
• The main tool used in the course is Flowise AI (a no-code platform for building chat flows and agent workflows). Detailed instructions for installing and accessing Flowise AI are provided in the course.

Course Topics

During the course, students will work with Flowise AI, a no-code tool that simplifies the creation of complex agent workflows.

Key course topics include:

Defining AI Agents

AI agents are LLM-based systems that perform tasks on behalf of the user. They are particularly useful for solving complex problems that require planning, access to tools, and memory.

Agent Components

AI agents include tools, memory, and planning for task execution. The course covers key components and how to use them effectively.

ReAct Agent

Students will learn about the concept of ReAct, which enables AI agents to analyze, respond to, and improve outcomes. The creation of a simple ReAct agent will be explored.

Agent Workflows

AI agents can be used in scientific research, programming, marketing, content design, and planning. This section covers how LLMs serve as the "brain" of the agent.

Flowise AI and Agent Workflows

Flowise AI provides a powerful toolkit for creating advanced AI agents by integrating the capabilities of LangChain and LlamaIndex. Students will create a search agent that retrieves current information from the internet.

Web Scraping Agent

An example agent will be examined that collects and analyzes data from the internet using search and information extraction tools.

Multi-agent Systems

The course demonstrates how different AI agents can interact and perform specialized tasks. For example, a copywriter agent system for marketing will be discussed.

Hierarchical Agents

The structure of supervisor/worker agents will be examined, where supervisor agents interact with worker agents to solve complex problems. A course scheduler agent will be developed as an example.

After completing the course, you will be able to independently develop AI agents for task automation in various fields without the need for programming.