-
Anthropic Courses — Personal notes on Official Anthropic courses. Recommended courses.
-
Quick Study List — Key skills and knowledge points organized by exam domain
-
Anti-Patterns — Common exam anti-patterns and what to avoid in each domain
-
Few-Shot vs Prompting Strategies — Decision framework for when to use few-shot examples vs explicit criteria, tool_use, or task decomposition (Domain 4)
-
Follow Epam Learn course Claude Certification: Claude Certified Architect – Foundations
- Certification Preparation Questions — Exam insights and community feedback from EPAM Leap (External)
- Guide to Preparing for the CCAF — Session preparation guide and study notes
- My Exam Feedback — Personal notes and topic breakdown from a real exam attempt
- Personal Notes — Key learnings, patterns, and insights from study sessions
- General Resources — Curated links to courses, official docs, and domain references
- Test Exam Notes — Practice exam questions notes Exam 1 · Exam 2
- Certification Exam Guide — Official exam guide covering domains, weights, and question format
- The Architect's Playbook — Anthropic's reference playbook for architecting Claude-based solutions
- AI Fluency Summary — One-page summary of core AI fluency concepts
- AI Fluency Vocabulary Cheat Sheet — Key terminology and definitions for AI and Claude concepts
The most important document you must study top to bottom is the Certification Exam Guide. All answers can be traced and supported by that information.
Recommended study path:
- 📖 Read the Certification Exam Guide cover to cover — understand all 5 domains and their weights
- 🎓 Complete the Anthropic courses — build hands-on familiarity with Claude tools and patterns
- 📝 Take a practice exam — identify knowledge gaps per domain
- 🔍 Deep-dive your weak domains using the Quick Study List and Anti-Patterns
- 🔄 Repeat practice exams until consistently scoring above 900
- Scenario Structure: 4 random scenarios selected from 6 possible scenarios
- Passing Score: 720 out of 1000 points
- Question Style: Real-world implementation scenario
The exam covers 5 core domains with the following weight distribution:
| Domain | Topic | Weight |
|---|---|---|
| 1 | Agentic Architecture & Orchestration | 27% |
| 2 | Tool Design & MCP Integration | 18% |
| 3 | Claude Code Configuration & Workflows | 20% |
| 4 | Prompt Engineering & Structured Output | 20% |
| 5 | Context Management & Reliability | 15% |
Total: 100%
Key Topics:
- Agentic loop mechanics and stop conditions
- Multi-agent orchestration patterns (hub-and-spoke)
- Subagent configuration and context isolation
- Workflow enforcement (hooks vs prompts)
- Task decomposition strategies
- Session management
Key Topics:
- Tool description best practices
- Error response handling
- Tool distribution and scoping
- MCP resources vs tools
- Built-in tools (Grep, Glob, Read, Write, Edit)
Key Topics:
- CLAUDE.md hierarchy (User → Project → Directory)
- Custom commands and skills
- Path-specific rules
- Plan mode vs direct execution
- Iterative refinement patterns
- CI/CD integration
Key Topics:
- Explicit criteria design
- Few-shot prompting techniques
- Structured output with tool_use
- Validation and retry strategies
- Batch processing
Key Topics:
- Context preservation strategies
- Escalation patterns
- Error propagation
- Large codebase exploration
- Human review and confidence calibration
- Information provenance
Good luck on your certification journey! 🚀
Last Updated: Study Guide Version - Foundations Certification