Learning path
AI Agents
Understand agent orchestration, tool use, task execution, and reliable automation patterns.
An AI agent does more than answer a chat. It plans the next step, uses a tool, checks the result, and keeps going until the job is done — or until it should stop. This path covers that loop: chatbots, document retrieval, planning, guardrails, and coordinating more than one agent.
Start here if you already write prompts and want systems that finish work. You do not need to be a software engineer. The early courses stay practical; later ones go deeper into reliability and multi-step runs.
What you will learn
Plan and act
Break a request into steps, choose a tool, and run the next action instead of stopping at a single reply.
Use your own data
Connect files and search so the agent answers from your documents instead of guessing from general knowledge.
Keep runs reliable
Add checks, limits, and recovery so an agent can fail a step without derailing the whole task.
How the path is sequenced
Take these in order if agent systems are new. Skip the first course if you already run a chatbot or custom GPT.
- Ship a simple chatbot — instructions, answers, and a clear job for the assistant.
- Point it at your files — retrieval so replies come from your process, not generic advice.
- Add planning and tools — multi-step agents that act, then check what happened.
- Harden the run — guardrails, recovery, and more than one agent on a larger job.
Recommended courses
These catalog courses belong on this path. Open any one for duration, price, and enrollment.
Start here
Beginner · $7 · 3.5 hours
Build a Simple AI Chatbot Without Coding
Create a basic chatbot that answers questions using your own instructions.
Intermediate · $20 · 10 hours
Retrieval Basics: Make AI Use Your Own Documents
Learn how AI can search your files instead of guessing from general knowledge.
Intermediate · $24 · 12 hours
Custom GPTs for Your Business Workflows
Build a private assistant trained on your process, not generic internet advice.
Core skills
Intermediate · $31 · 15.5 hours
AI Agents That Plan, Act, and Finish Tasks
Go beyond chat and learn how agents break work into steps.
Intermediate · $32 · 16 hours
Build a RAG Pipeline from Scratch
Connect documents, search, and an LLM into one working answer system.
Advanced · $37 · 18.5 hours
AI Security and Guardrails That Hold Up
Reduce leaks, jailbreaks, and unsafe answers in real apps.
Go further
Advanced · $42 · 21 hours
Multi-Agent Orchestration for Complex Work
Coordinate several agents so they divide work instead of colliding.
Intermediate · $49 · 10 hours
Build AI Apps with Real APIs
Ship chat, summaries, and structured AI features into a working product.
Advanced · $79 · 12 hours
AI Agents and Automation That Finish Work
Design multi-step agents that plan, call tools, and recover from mistakes.
What a reliable agent needs
A single clever prompt is not an agent. Across this path you will practice putting these parts in place:
- Goal — the outcome the run should reach, in language a person would accept as “done.”
- Tools — search, documents, APIs, or forms the agent is allowed to call.
- Memory — what to keep from earlier steps so the next action is not starting from zero.
- Checks — how to tell a good result from a guess, a loop, or an unsafe action.
- Stop conditions — when to ask a person, retry, or end the run.
Who this path is for
This sequence is a fit if you already use chat tools and now need work to move without you copying every step by hand. It also suits product and operations people who want agents that stay inside a defined job.
If you are still learning how to write a clear prompt, start with the Prompt Engineering path. If your next step is connecting models to everyday process flows and APIs, continue to AI Automation.
Begin with a chatbot that follows your instructions
The first course is $7. Build a simple assistant, then move into planning, retrieval, and agents that finish tasks.