AI Coder: Claude Code & Coding Agents

What's inside

1. Level 1: Mental Models of AI-Assisted EngineeringFree
   Understand how coding agents think, where context windows break down, and how to shift from manual coding to specification-driven development.
2. Level 2: Inline Completion Mastery: Copilot & CodexFree
   Leverage GitHub Copilot and OpenAI Codex effectively by steering inline suggestions through prompt comments and active context cues.
3. Level 3: Context-Aware Editing with Cursor IDE
   Harness Cursor's codebase indexing, composer mode, and custom system rules to generate multi-file edits safely.
4. Level 4: The Vibe Coding Methodology & Safety Guardrails
   Establish a fast, test-driven vibe coding loop while setting up deterministic git guardrails and linting checks.
5. Level 5: Getting Started with Claude Code CLI
   Install, authenticate, and configure Anthropic's Claude Code CLI inside real terminal development environments.
6. Level 6: Claude Code Permissions & Execution Sandbox
   Configure execution permissions, auto-approve boundaries, and sandboxing rules for shell commands and file mutations.
7. Level 7: Context Compaction & Token Budgeting
   Manage context windows during extended Claude Code sessions using manual compaction, slash commands, and token budgeting.
8. Level 8: Sub-Agent Task Delegation in Claude Code
   Orchestrate complex engineering tasks by letting Claude Code spawn and manage specialized sub-agents for research and implementation.
9. Level 9: End-to-End Refactoring with Claude Code
   Execute a full-stack architectural refactor across multiple files and test suites using Claude Code terminal workflows.
10. Level 10: Introduction to the Model Context Protocol (MCP)
    Learn the architecture of MCP and connect coding agents to external filesystems, databases, and local runtimes.
11. Level 11: Extending Agents with Custom MCP Servers
    Build and deploy a custom MCP server in TypeScript/Python to expose proprietary internal APIs and debuggers to your agents.
12. Level 12: Database & API Tooling via MCP
    Empower coding agents to safely inspect live schemas, run read-only queries, and mock external API integrations.
13. Level 13: Browser Automation & Visual Feedback with MCP
    Integrate headless browser MCP servers to give coding agents visual feedback, DOM inspection, and UI debugging capabilities.
14. Level 14: OpenCode: Open-Source Terminal Agents
    Run and customize OpenCode using local and open-weights models for privacy-sensitive and offline code generation.
15. Level 15: Antigravity: Autonomous Issue Resolution
    Deploy Antigravity to monitor repositories, ingest bug reports, reproduce failures, and generate patch pull requests.
16. Level 16: OpenClaw: Multi-Agent Swarm Architectures
    Architect multi-agent engineering swarms using OpenClaw to divide roles into Product Manager, Architect, Coder, and QA.
17. Level 17: Self-Healing Test Suites & Automated Debugging
    Create automated feedback loops where failing integration tests trigger autonomous multi-agent root-cause analysis and repair.
18. Level 18: Capstone: Production Vibe Coding Orchestration Exam
    Plan, scaffold, implement, test, and ship a complete full-stack feature from scratch using an orchestrated multi-agent harness.

Access

The first 2 levels are free with a free account. Every level, the podcast edition and the AI tutor come with All Access at £4.99/month or any Creator plan — see pricing.

Start free →