Building AI Agents: From Chatbots to Autonomous Workflows

A more technical, hands-on session where participants design and build agentic AI systems that can use tools, make decisions, and chain tasks together.

About This Course

AI agents are the next frontier. Unlike basic chatbots that answer questions, agents can use tools, browse the web, execute code, make decisions, and complete multi-step tasks autonomously. This course teaches you to build them.

Across three sessions, you will go from understanding agent architecture to building and deploying your own AI agents. We cover the core concepts — tool use, memory, planning, and orchestration — and put them into practice with hands-on projects using Python and modern agent frameworks.

This is the most advanced course in the series and the most exciting. You will build AI systems that can actually do things in the real world, not just generate text.

Who This Course Is For

Developers and aspiring developers who want to build cutting-edge AI applications
Students with some programming experience who want to work in AI
Technical professionals who want hands-on experience with agentic AI
Graduates of the Coding with AI course who want the next challenge
Anyone with basic Python skills who is fascinated by autonomous AI systems

What You Will Walk Away With

A working AI agent you designed and built, deployed and functional. Understanding of the agent architecture patterns used by the leading AI labs. The skills to continue building increasingly sophisticated agents independently.

What You'll Learn

Explain the architecture of AI agents: perception, reasoning, planning, and action
Build AI agents that can use external tools including web search, code execution, and APIs
Implement memory systems that let agents maintain context across interactions
Design multi-step agent workflows with decision-making and error recovery
Use modern agent frameworks including LangChain, CrewAI, or similar tools
Build a multi-agent system where specialized agents collaborate on tasks
Deploy a functional AI agent that solves a real-world problem
Evaluate agent performance and debug agent behavior

Session Breakdown

Session 1: Agent Architecture and Your First Agent

What makes an agent different from a chatbot: the sense-think-act loop
Agent architecture: tool use, memory, planning, and orchestration
The agent landscape: LangChain, CrewAI, AutoGen, and direct API approaches
Hands-on: Build your first agent — an AI that can search the web and summarize results
Tool use: giving your agent the ability to interact with the world
Prompt engineering for agents: system prompts, instructions, and guardrails
Testing and debugging agent behavior

Session 2: Memory, Planning, and Multi-Step Workflows

Memory systems: giving agents the ability to remember and learn
Short-term vs. long-term memory: conversation history and persistent knowledge
Planning: how agents break complex tasks into steps
Hands-on: Build an agent with memory that improves over multiple interactions
Multi-step workflows: chaining actions for complex tasks
Error handling and recovery: what happens when agents fail
Guardrails and safety: keeping agents from going off the rails

Session 3: Multi-Agent Systems and Your Capstone

Multi-agent systems: when one agent is not enough
Agent collaboration patterns: delegation, specialization, and orchestration
Hands-on: Build a multi-agent system — e.g. a research team with analyst, writer, and reviewer agents
Capstone project: design, build, and deploy your own agent for a real task
Demo time: show the class your agent in action
The future of agents: where this technology is heading
Continuing your journey: resources, communities, and next steps

Prerequisites

Basic Python programming experience is required — you should be comfortable writing functions, using loops, and working with APIs. Completing Coding with AI or having equivalent experience is strongly recommended. You should also have a basic understanding of AI concepts from any of the earlier courses.

What to Bring

A laptop with Python 3.10+ installed, a code editor like VS Code, and the ability to install Python packages via pip. An OpenAI API key or Anthropic API key is required — instructions for obtaining free-tier access will be sent before class. Pre-setup instructions will be provided one week before the course.

$249

per person

No sessions currently scheduled

Bundle Discount

Save up to 30% when enrolling in multiple courses. Use code TRACK_30 for the full track.