Free & Open Source Learning Path

Become an
AI Coding
Champion.

A structured 10-level path from first setup to running parallel agent teams. Stop guessing. Start building with intention.

10 Levels · 116 Actions · Track Progress · Always Free
Claude Code Gemini CLI GitHub Copilot
Start the Roadmap View on GitHub

The Path

10 levels. Zero fluff.

Every level builds on the last. Each one has clear actions, explanations, and commands — not just theory.

Level 0 5 actions
Foundation: AI Resources Repository
One git repo that survives machine changes and holds every config, skill, hook, and learning.
 Level 1 9 actions
Context Engineering
Each tool's default behavior matches your preferences without prompting on every session.
 Level 2 9 actions
Compounding Engineering
Each cycle of work improves future cycles. 80% planning and review, 20% execution.
 Level 3 12 actions
Skills and Extensions
The agent gains specialized capabilities without permanently consuming context.
 Level 4 10 actions
MCP Servers
The agent can search the web and reach external systems. MCP context cost is managed.
 Level 5 6 actions
Prompting Discipline
Better first responses and fewer correction loops.
 Level 6 13 actions
Token Management and Observability
Know what the context window contains. Stop burning tokens on noise.
 Level 7 19 actions
Hooks and Persistent Memory
The agent detects its own errors without prompting. Sessions start with context from previous ones.
 Level 8 10 actions
Remote Setup
An always-on agent server reachable from anywhere. No laptop dependency.
 Level 9 18 actions
Parallel Agents
Multiple agents working on independent tasks simultaneously.

Ready to start?

Open the roadmap, tick off Level 0, and you're already ahead.

Open the Roadmap View Cheatsheet