Fundamentals of Building AI Agents

Overview

Google, IBM & Meta Certificates — All 10,000+ Courses at 40% Off

One annual plan covers every course and certificate on Coursera. 40% off for a limited time.

Are you ready to build AI that thinks, acts, and gets things done? In this course, you’ll learn how to design agents that go beyond language generation to reason, take action, and tackle real-world tasks using tools and data.  During the course, you'll explore the foundations of tool calling and chaining with LangChain. You’ll discover how to extend the capabilities of Large Language Models (LLMs) by connecting them with calculators, code, and external data sources. You'll learn how LLMs trigger tool use through LangChain Expression Language (LCEL) and look at manual tool calling for greater control and accuracy. Plus, you’ll explore built-in agents that can analyze data, create visualizations, and run SQL queries using natural language.  To get the most from this course, we recommend that you have Python programming skills, a basic understanding of LangChain, and familiarity with core AI concepts.  Whether you're building a chatbot or a smart assistant, if you’re looking to build the skills to create dynamic, intelligent, and goal-oriented AI systems, enroll today!

Syllabus

Foundations of Tool Calling and Chaining

This module introduces AI agents and explains how they differ from traditional large language model workflows. You will explore how agents use reasoning, tools, and memory to perform multi-step tasks and real-world interactions. The module also covers tool calling and chaining in LangChain, including how to design and integrate custom and pre-built tools. Through hands-on practice, you will begin building AI agents capable of executing structured, goal-oriented workflows.

LCEL and Manual Tool Calling in LangChain

This module focuses on building structured workflows using LangChain Expression Language (LCEL) and implementing manual tool calling for greater control. You will learn how to construct chains, extract tool inputs from LLM outputs, and validate and execute tool calls effectively. The module also explores how to bind custom tools to models and manage tool invocation for accuracy, safety, and cost efficiency. Through labs, you will develop agents that combine automated reasoning with controlled execution.

Using Built-in Agents in LangChain

This module explores the use of pre-built agents in LangChain for data analysis and database interactions. You will learn how to configure and use DataFrame and SQL agents to process natural language queries and generate insights. The module also demonstrates how these agents translate conversational input into structured operations for visualization and data retrieval. Through hands-on labs, you will build AI-powered applications that enable intuitive interaction with data systems.