Learning path
Build an AI Operating System
1 h 30 min — 2 courses — 12 readings
At a glance
- Who it is for
- Builders and practitioners who want a working AI environment of their own.
- By the end
- You can assemble a working AI environment with context, durable memory, tools, permissions, orchestration and checks that prove it works.
- Starts at
- Getting started. No experience needed.
- Highest level
- Level 4 of 6, AI operating system that learns.
What you will be able to do
- Explain what an AI keeps in context, what resets and how memory and retrieval fill the gap.
- Set up a local AI lab and an orchestration layer with tools you already have.
- Turn a written rule into an enforced control and a CI gate.
- Run the system without starving everything around it.
Courses in this path
Course 1 of 2
Foundations: how an AI keeps and uses context
40 min — 6 readings
In this course you will:
- Tell context, context window, memory and retrieval apart.
- Give an agent memory that survives a new session.
Stage 1: Getting started: contextYou are here
- Reading 1 of 12 — your next stepWhat Is Actually Inside an AI’s Context, and What Makes It Reset?
Tell what is inside an AI’s context apart from what makes it reset.
- Reading 2 of 12What Is a Context Window and Why Does It Limit What an AI Can Remember?
Understand what a context window is and why it limits what an AI can remember.
Stage 2: Memory and retrieval
- Reading 3 of 12How Does an AI Agent Remember Things Between Conversations?
See how an AI agent keeps things between conversations.
- Reading 4 of 12How do I give an AI agent memory that survives a new session?
Build and test a durable instruction that a later session can retrieve.
- Reading 5 of 12What Is RAG (Retrieval-Augmented Generation) and How Does It Work?
See what retrieval-augmented generation is and how retrieval fills the gap memory leaves.
Stage 3: Taking action
- Reading 6 of 12What Is Tool Calling and How Does It Let an AI Take Real Actions?
See how tool calling lets an AI take real actions.
Course 2 of 2
Build and run your own
50 min — 6 readings
In this course you will:
- Prepare hardware and a free local lab.
- Add an orchestration layer, an enforced control and a CI gate.
Stage 1: Set up your lab
- Reading 7 of 12How Do You Enable Your Hardware for Local AI Before You Install Anything?
Prepare your hardware for local AI before you install anything.
- Reading 8 of 12How to Build a Free AI Lab Using Tools You Already Have
Build a free AI lab using tools you already have.
- Reading 9 of 12How Do You Set Up an Orchestration Layer for Local AI Agents?
Set up an orchestration layer for local agents.
Stage 2: Gates and controls
- Reading 10 of 12How Do I Turn a Rule Into an Enforced Control?
Turn an explicit rule into a check that denies the prohibited state.
- Reading 11 of 12What Is a CI Gate and Why Does It Block Bad Code From Merging?
See what a CI gate is and why it blocks bad code from merging.
Stage 3: Run it well
- Reading 12 of 12Running AI Well Means Managing Everything Around It
Run the system without starving everything around it.