Claude Certified Architect – Foundations: My Certification Journey and Preparation Path

Sun, 24 May 2026 00:00:00 +0000

On May 20, 2026 I passed the Claude Certified Architect – Foundations exam. The certificate landed in my inbox shortly after, with a verification URL on Skilljar and a six-month validity window. The credential itself is just a PNG. The journey behind it is what I want to talk about, because the exam is not what most people expect when they hear "AI certification." It is not a quiz about prompt templates and parameter names. It is a brutally practical test of whether you can architect systems with Claude — and most of the questions are designed in a way that will not reward you for memorizing documentation.

Claude Certified Architect — Foundations

Wed, 20 May 2026 00:00:00 +0000

Issued by: Anthropic, PBC Credential: Claude Certified Architect – Foundations Certificate No.: v9axz9t5re5n Valid: May 20, 2026 – Nov. 20, 2026 Verify: verify.skilljar.com/c/v9axz9t5re5n

The Claude Certified Architect – Foundations is Anthropic’s official credential for solution architects who design and ship production applications with Claude. The exam validates practical judgment — not just conceptual knowledge — across the four core technologies that build production-grade Claude systems:

Claude Agent SDK — agentic loops, subagent orchestration via the Task tool, hooks (PostToolUse), stop_reason handling, and session management with fork_session / --resume.
Claude Code — CLAUDE.md configuration hierarchy, .claude/rules/ with YAML path-scoped glob patterns, project-scoped slash commands and skills (context: fork, allowed-tools), plan mode, and CI integration with -p / --output-format json.
Model Context Protocol (MCP) — tool interface design, structured error responses (isError, errorCategory, isRetryable), tool distribution and tool_choice configuration, and server scoping (.mcp.json vs ~/.claude.json).
Claude API — tool_use with JSON schemas for guaranteed structured output, tool_choice options (auto / any / forced), the Message Batches API, and context window management strategies.

Exam Domains and Weightings

Domain 1: Agentic Architecture & Orchestration (27%)
Domain 2: Tool Design & MCP Integration (18%)
Domain 3: Claude Code Configuration & Workflows (20%)
Domain 4: Prompt Engineering & Structured Output (20%)
Domain 5: Context Management & Reliability (15%)

Scored 100–1,000 (passing 720), the exam draws from six production scenarios — customer support agents, Claude Code in development, multi-agent research pipelines, developer productivity tooling, CI/CD review automation, and structured data extraction — and tests architectural tradeoffs in each.

MCP on Vadzim Belski — AI Research & Engineering

Claude Certified Architect – Foundations: My Certification Journey and Preparation Path

Claude Certified Architect — Foundations

Exam Domains and Weightings