Claude Training Resources

Choose your learning path

Start with Claude Foundations, go deeper with the restored advanced agent labs, or test readiness with the Cloud Architect Practice Exam.

View Paths → Quick Reference

Learning Paths

Pick the track that matches your goal

Claude Foundations

Twelve hands-on modules covering the Claude platform end to end, from environment setup and agentic loops through MCP, governance, reliability, and the final orchestrator project.

Claude Advanced Labs - Agents

A deeper agent-building track with setup, advanced modules, self-driven Jupyter labs, answer keys, and the Enterprise Prospecting & Outreach Orchestrator capstone.

✓

Cloud Architect Practice Exam

Randomized question sets across all five domains with score breakdowns, weak-topic detection, and a NotebookLM study prompt for targeted review.

Companion Resource

Use NotebookLM alongside this lab

Claude API Development and Integration Guide

The source material for this lab lives in a NotebookLM notebook. Use it to ask follow-up questions, explore topics in more depth, or get a plain-English explanation of any concept while working through the modules.

Open in NotebookLM

Claude Foundations

Foundations modules

Each module builds on the last. By the end you'll have a working blueprint for an autonomous AI marketing agent, and a solid grasp of every feature the exam tests.

Dev Environment & Foundations

Install VS Code, set up Jupyter notebooks, and connect the Anthropic Python SDK to verify your environment before Module 2.

Agentic Loops & SDK Foundations

Master the Messages API stateless design (including how tool_result blocks re-enter history), set up the Agent SDK, identify ZDR-eligible features, and control conversation flow with the agentic loop stop_reason pattern.

Tool Design & Tool Choice Strategy

Write disambiguating tool descriptions to prevent misrouting, force prerequisite calls with tool_choice, and enforce strict JSON schemas on tool inputs.

Multi-Agent Orchestration (Agent SDK)

Build a hub-and-spoke coordinator with the Agent SDK Task tool, spawn subagents in parallel, and pass context explicitly across their isolated histories.

Scaling & Batch Constraints

Contrast synchronous loops with async batches, internalize the no-multi-turn-tools constraint, and resubmit only failed records by custom_id at a 50% discount.

Claude Code & Skills

Explore large codebases with Grep and Glob, use deterministic Edit fallbacks, compare fixes with forked sessions, and govern repository standards with CLAUDE.md and path-scoped rules.