Become a Claude Certified Architect

A free, visual, build-it-to-learn-it site for Anthropic's CCA-F exam. Explainers you drive, quizzes that talk back, code you run (Colab or local), and projects you can put on a résumé.

← a real agent loop, running. You'll build one of these.
0
interactive explainers
0
quiz questions
0
runnable exercises
0
portfolio builds
0
exam traps decoded

🚀 New here? Play before you read.

In The Agent Hangar, an AI agent is a spaceship: master each part in a ~1-minute mini-game (the model = reasoning core, tools = the robotic arm, context = cargo hold, budget = fuel, guardrails = shields), assemble the ship, then fly missions where every decision point is a real exam judgment call — wrong picks cost hull and teach the trap. Then the explainers, quizzes, and code make it stick.

Enter the hangar →

The path — nine steps, in order

Each step is learn (drive the explainers) → build (run/ship something) → prove (drills). The sequence is deliberate: you need prompts and tools before agents make sense. Full pacing in the 6-week study plan.

  1. Foundations — what even is the API?
    CLI vs API vs SDK untangled · how claude.ai & VS Code map to it · get your key · first calls. Then run the 10 hands-on exercises (Colab or local).
  2. Prompt engineering & structured output 20%
    Prompt anatomy · few-shot · JSON schemas & nullable fields · eval-driven changes. Build proj-05: extraction pipeline.
  3. Tool design & MCP 18%
    The description IS the API · structured errors · MCP host/client/server. Build proj-03: your own MCP server.
  4. Agentic architecture 27% — the big one
    The loop · workflows vs agents · multi-agent orchestration · guardrails. Build proj-01: support agent and proj-02: research system.
  5. Claude Code configuration 20%
    CLAUDE.md hierarchy · Grep vs Glob · skills/rules/hooks · headless CI. Build proj-04: Claude Code in CI/CD.
  6. Context management & reliability 15%
    Context window · prompt caching · compaction & memory · escalation. Build proj-06: context engine.
  7. Drill every domain — 🎯 quiz mode
    Interactive scenario MCQs with instant feedback, "why the distractor is wrong" explanations, and best-score tracking. (Prefer reading? The printable drills have the same questions.)
  8. Capstone: Enterprise Agent Platform
    Stitch the six builds into one governed system — the portfolio piece that says architect.
  9. Timed mock exam → book the real one
    Closed-book, mixed scenarios. Score ≥10/12, review misses against the 14 traps, then book.

The five exam domains

Practice anywhere — three ways to run the code

Every exercise and project can be done with zero-install (Colab) or on your machine. Pick the lane that matches your setup; the code is identical.

☁️ Google Colab — zero install

Run in the browser

  1. Open notebooks/foundations-api.ipynb in Colab (File → Upload, or open it straight from GitHub once this repo is pushed).
  2. Run the first cell — it installs the SDK and asks for your API key with a hidden prompt (get a key).
  3. Run the 10 exercises top to bottom. Total cost: a few cents on Haiku.
🖥️ VS Code — Copilot Chat or Claude Code

Learn inside your editor

  1. Create a folder, `pip install anthropic`, export your key, and work through the exercises as .py files.
  2. Stuck? Ask Copilot Chat (pick a Claude model) or the Claude Code extension to explain the error — you're literally using the Domain 3 material while studying it.
  3. This explainer maps every editor feature to the API concept underneath.
⌨️ Terminal — Claude Code CLI / claude.ai

Interactive & agentic

  1. Domain 3 is hands-on by nature: install the Claude Code CLI and do the CLAUDE.md / hooks / headless exercises in a real repo.
  2. Use claude.ai to test prompt-engineering patterns interactively before coding them.
  3. Projects 1–6 run from any of the three lanes — each brief says what it needs.

Which model, when — with real numbers

"Use the cheapest capable model" only sticks when you've seen the actual math. The model-selection explainer now has a cost calculator: dial in your volume and token sizes and watch the monthly bill for Haiku vs Sonnet vs Opus — e.g. classifying 1M support tickets is ≈ $650 on Haiku vs ≈ $9,750 on Opus for near-identical accuracy on that task. Three worked case studies show how an architect actually decides.

What you'll ship (the résumé part)

proj-01

Customer-Support Agent

Tools + guardrails + human approval + escalation; evaluated at 80%+ resolution.

proj-02

Multi-Agent Research System

Lead + isolated subagents, provenance, partial-failure handling.

proj-03

Custom MCP Server

Publishable server any Claude host can install — tools, resources, prompts.

proj-04

Claude Code in CI/CD

Headless PR review + test-gen Action, hooks, layered CLAUDE.md.

proj-05

Extraction Pipeline

Schema contracts, nullable handling, Batch API, LLM-as-judge evals.

proj-06

Context Engine

Compaction + verbatim memory + caching; 100-turn sessions at flat cost.

⭐ capstone

Enterprise Agent Platform

All six under one governance spine: routing, cost, guardrails, evals, audit.