AI Agents 101
Chatbots answer questions. Agents do things — search the web, call tools, remember context, and work through multi-step tasks. This course demystifies how agents work under the hood and walks you through building your first real one, no PhD required.
What you'll work through
Click a module to see its lessons. You'll build with real agent frameworks as you learn the concepts.
Module 1 — From chatbots to agents
- What makes an agent an agent: autonomy, tools, and goals Clear definitions without the hype.
- The agent loop: perceive, reason, act, repeat Exercise: trace an agent's loop step-by-step on paper.
- Where agents shine — and where they flop Honest tour of strengths, costs, and failure modes.
Module 2 — Tools: giving agents hands
- How tool calling works under the hood Function schemas, arguments, and results — demystified.
- Designing good tools: narrow, named, and documented Exercise: spec three tools for a travel-planning agent.
- Safety basics: what an agent should never be allowed to do Permissions, confirmations, and blast-radius thinking.
Module 3 — Memory and context
- Short-term memory: conversation history and context windows Why agents "forget" and how to manage it.
- Long-term memory: notes, files, and vector stores Exercise: design a memory scheme for a research agent.
- Retrieval: letting agents look things up RAG in plain English — documents in, answers out.
Module 4 — Build 1: Your research agent
- Planning the build: goal, tools, and success criteria Scope a genuinely useful agent before writing anything.
- Wiring search and summarization tools Hands-on: connect web search to your agent.
- The reasoning loop in action Watch it plan, search, read, and synthesize.
- Testing and hardening your agent Break it on purpose, then fix what you find.
Module 5 — Build 2: A personal task agent
- Designing an agent around your own workflow Pick a repetitive task from your real life.
- Build 2: multi-step task automation Guided build with checkpoints and troubleshooting tips.
- Next steps: multi-agent systems and production Bridges into the premium Agents in Production course.
What you'll be able to do
Prerequisites
Prompt Engineering Basics recommended (or equivalent comfort with AI chat tools). Basic computer confidence helps; no programming background required for the concepts, though the builds include optional code walkthroughs.
Instructor
HelpLearnAI curriculum team — designed by our in-house curriculum team with an emphasis on building real, working agents.
Build your first agent.
Join the waitlist and we'll notify you the moment AI Agents 101 opens for enrollment.
Enroll — join waitlist