Claude Certified Developer - Foundations Certification

Domain 1 Agents and Workflows (14.7%)

• Apply agent and workflow architecture patterns and decide when to use a workflow versus an agent
• Construct Claude agents with the Claude Agent SDK, custom agent loops and harnesses, and hooks for deterministic actions
• Choose managed deployment models (self-hosted vs Anthropic-hosted) and design manager/subagent hierarchies
• Apply agent design patterns (tool-use loops, sub-agents, memory, context-window management) and frameworks such as Strands, LangGraph and PydanticAI

Domain 2 Applications and Integration (33.1%)

• Translate business requirements into functional and infrastructure requirements and manage the systems life cycle
• Apply Claude API mechanics (messages, tools, streaming, vision, thinking, caching, batch API, realtime vs batch tradeoffs)
• Apply software engineering foundations (REST APIs, JSON, async programming, version control, code review, refactoring)
• Design Claude applications across interfaces (Claude Code, Desktop, claude.ai, API, SDKs) with sound schema design and session hygiene
• Manage configuration with CLAUDE.md files, settings.json, model version pinning, prompt versioning and plugin dependencies

Domain 3 Claude Code (3.1%)

• Operate Claude Code core components (Rules, Skills, Commands, Agents, Agent Memory)
• Use session management, built-in and custom slash commands, headless, streaming and auto modes
• Apply the CLAUDE.md hierarchy, repository initialization and settings.json configuration

Domain 4 Eval, Testing and Debugging (2.6%)

• Debug and handle errors in Claude applications (error type identification, recovery strategy selection)
• Use trace analysis to identify failure modes
• Isolate problem origin between the integration layer and model output

Domain 5 Model Selection and Optimization (16.8%)

• Apply LLM fundamentals (tokens, context windows, sampling, non-determinism, next-token generation) and thinking modes (fast, extended, adaptive, effort levels)
• Apply technical fundamentals of AI application development (SDKs wrapping REST APIs, websockets)
• Select models by capability and tradeoffs (Opus vs Sonnet vs Haiku; quality, latency, cost; behavior changes across releases)
• Manage cost and tokens (usage tracking, cost modeling, prompt caching, cache check-pointing)

Domain 6 Prompt and Context Engineering (11.0%)

• Engineer context (context-window management, preventing context drift and bloat via tool-output pruning and compaction, context isolation through subagents)
• Engineer prompts (instruction clarity, few-shot examples, system vs user placement, output constraints, iterative refinement, input sanitization)
• Handle output with structured-output patterns, response validation, defensive parsing and skepticism toward confident output

Domain 7 Security and Safety (8.1%)

• Apply AI application security (prompt-injection mitigation, jailbreak defense, untrusted-input handling, data-leakage and PII prevention, authentication, authorization, confidentiality and integrity)
• Apply guardrails and secure-by-design deployment (content policy, guardrail layering, least privilege)
• Leverage Claude hooks to prevent destructive actions
• Manage identity, secrets and API keys across development and production environments

Domain 8 Tools and MCPs (10.6%)

• Implement tools and function calling (tool description writing, error handling, agentic harness dispatch, client-side vs server-side tools, approval patterns)
• Weigh tradeoffs across built-in tools, custom tools, Skills and MCPs
• Develop MCP servers (server authoring, deployment, MCP resources, tools and prompts)
• Apply MCP communication patterns (stdio, sockets, client vs server)