Stage 02 of 03 · Builder
OpenClaw: Build Your Own AI Agent
Your AI. Your tools. Your memory.
What students learn
Most students only know AI as a chat window. This course teaches them to own one. Students set up OpenClaw — an open-source AI agent toolkit — on their own computer, learn how persistent memory works, give the agent custom tools (browser, files, messaging), and watch it accomplish tasks no chatbot can. Beyond the technical chops they walk away with the most important AI literacy fact of this decade: a personal agent that remembers you and operates your tools is a fundamentally different thing than ChatGPT — and being on the building side of that shift is a career-shaping advantage.
What an AI agent actually is (a loop, not a chat). How agents differ from chatbots. How memory persists across sessions and conversations. How tools extend what an AI can do for you. The mental model behind why personal agents will quietly replace generic chat over the next few years — and how to be on the building side of that shift, not just the using side.
Module-by-module
-
01
Chat vs. agent
Install and run OpenClaw locally; watch the agent loop run.
-
02
Persistent memory
Save and recall facts across sessions. Agent remembers you.
-
03
Tool basics
Give the agent a Python tool you wrote. Agent calls it on demand.
-
04
Browser as a tool
Agent that researches a topic on the open web for you.
-
05
Messaging as a tool
Agent that pings you on Telegram or Discord when something happens.
-
06
Files as a tool
Agent that organizes a folder, summarizes a document, edits a note.
-
07
Mini-project, part 1
Pick a useful personal job for an agent to do. Working v1.
-
08
Mini-project, part 2 + demo
Polish, deploy, share. A personal agent the student uses every day.
What they make
A working personal AI agent the student uses every day — with memory, custom tools, and a real job to do.
Walk away with
The instinct that AI is something you build with, not something you submit prompts to. A working agent system. The mental model of agency, memory, and tools that the next decade of AI products will be built on.
Learn on your own pace
This course can be self-studied end-to-end using these vetted free resources. Work through them at your own pace — alone, or with a parent alongside.
-
🥇
OpenClaw
The open-source agent toolkit used by the course. Memory, tools, and a clean agent loop built for learning.
-
🥈
Anthropic — Building Effective Agents
The seminal essay on what makes an agent good. Required reading for anyone building one.
-
🥉
Claude Code
A production AI agent that lives in your terminal. Try it; reverse-engineer how it works.
Honorable mentions
Next step
Innovator project that uses AI agents to solve a real problem.
Ready to enroll?
OpenClaw: Build Your Own AI Agent
Pick the course your student is curious about (or "not sure yet") and we’ll get back to you with next steps. Most students start in Explorer or Builder based on age and prior experience — we’ll help you find the right fit.