# Claude Certified Engineers > A software studio staffed by engineers certified through Anthropic's Claude Certification Program. We ship production Claude agents, MCP integrations, and eval harnesses. Fixed fee, your IP. ## Site - [Home](https://claudecertified.agency/): Studio overview, certifications, services, and pricing. - [Blog](https://claudecertified.agency/blog): Notes on production Claude systems. - [Certifications](https://claudecertified.agency/certifications): Anthropic Claude Certification Program exams we hold. - [Services](https://claudecertified.agency/services): Audit, production agent build, MCP integration, eval harness, enablement, embedded engineer. - [FAQ](https://claudecertified.agency/faq): Common questions about the studio and certifications. - [Process](https://claudecertified.agency/process): How engagements run from discovery to delivery. - [Bench](https://claudecertified.agency/bench): Certified engineer bench and staffing. - [Pricing](https://claudecertified.agency/#pricing): Fixed-fee engagements. ## Production agents - [Guardrails and prompt injection for tool-using Claude agents](https://claudecertified.agency/blog/guardrails-prompt-injection-claude-agents): A tool-using Claude agent without guardrails will be helpful in the wrong direction. Here is the refusal, injection, and review path we install before live traffic. - [Prompt caching and token budgets in production Claude](https://claudecertified.agency/blog/prompt-caching-token-budgets-production): Prompt caching cuts Claude spend when the stable prefix is actually stable. Here is how we set cache breakpoints, ceilings, and evals so cost is a number, not a surprise. - [LLM-as-judge rubrics you can defend in CI](https://claudecertified.agency/blog/llm-as-judge-rubrics-in-ci): LLM-as-judge is fine when the rubric is boring, structured, and checked by a human. Here is how we score Claude agents without turning CI into a $12 hobby. - [Put an eval harness on Claude before the agent goes live](https://claudecertified.agency/blog/eval-harness-for-claude-agents): Most Claude pilots stall because nobody can tell if quality got better. Here is the eval harness we install in week one, wired into CI, before the agent sees real traffic. - [Golden datasets for Claude agents](https://claudecertified.agency/blog/golden-datasets-for-claude-agents): A golden dataset for Claude is a named, versioned set of real cases with invariants, not a folder of happy-path chats. Here is how we build one that CI can fail. ## MCP and tools - [Connecting Claude to Salesforce or SAP without a catch-all tool](https://claudecertified.agency/blog/claude-salesforce-sap-mcp): "We connected Claude to Salesforce" usually means one mega-tool and a security exception. Here is how we ship named MCP verbs that survive a CRM of record. - [Human-in-the-loop queues for irreversible MCP actions](https://claudecertified.agency/blog/human-in-the-loop-mcp-review-queues): If Claude can refund, post, or export, it needs a review queue, not a braver system prompt. Here is how we design human checkpoints that agents actually wait on. - [Claude tool use vs MCP: when each is the right contract](https://claudecertified.agency/blog/claude-tool-use-vs-mcp): Tool use on the Messages API and MCP servers are not the same design. Here is how we choose which contract Claude should use for a given job. - [MCP server security review that clears InfoSec](https://claudecertified.agency/blog/mcp-server-security-review): An MCP server is a permissioned adapter, not a chatbot plugin. Here is the security review we run before Claude gets hands on Salesforce, SAP, or your internal APIs. - [Tool schemas Claude actually calls](https://claudecertified.agency/blog/mcp-tool-schemas-claude-calls): Claude skips, mis-fills, or over-calls tools when the schema is written for OpenAPI completeness instead of for a model. Here is how we design MCP tools that get used correctly. ## Claude Code - [Claude Code adoption metrics: measure team value without ranking engineers](https://claudecertified.agency/blog/claude-code-adoption-metrics-scorecard): Build a team-level Claude Code scorecard with clear denominators, delivery measures, privacy rules and a worked decision about rollout expansion. - [Claude Code Policy Exception Workflow: Approve and Audit Elevated Access](https://claudecertified.agency/blog/claude-code-policy-exception-workflow): A risk-tiered workflow for requesting, approving, verifying, expiring, and auditing Claude Code access exceptions without weakening the baseline. - [Claude Code enterprise rollout checklist: permissions, policy, and auditability](https://claudecertified.agency/blog/claude-code-enterprise-rollout-checklist): A staged Claude Code enterprise rollout checklist covering managed policy, tool permissions, MCP, gateways, auditability, evals, cost controls, and adoption. - [CI eval gates for Claude Code changes](https://claudecertified.agency/blog/ci-eval-gates-claude-code): If Claude Code can edit prompts and tools, CI must be able to fail those edits. Here is how we gate agentic coding the same way we gate application code. - [Certification enablement vs hiring another Claude engineer](https://claudecertified.agency/blog/certification-enablement-vs-hiring): Enablement trains your team on your repo. Hiring adds a headcount. Here is how we choose, and why the exams are the syllabus rather than trivia. - [CLAUDE.md that the team actually follows](https://claudecertified.agency/blog/claude-md-that-teams-follow): CLAUDE.md is repo policy for Claude Code, not a novel. Here is what we put in it so agentic coding stays inside your conventions after we leave. - [Repo conventions that keep Claude Code inside your architecture](https://claudecertified.agency/blog/agentic-coding-repo-conventions): Agentic coding fails when the repo has no rules the agent can follow. Here is the convention set we install so Claude Code PRs look like your team's, not like a weekend plugin. - [First PR in seven days with Claude Code in an existing repo](https://claudecertified.agency/blog/first-pr-claude-code-existing-repo): A certified engineer opens a pull request inside seven days because the repo is ready, not because the model is magic. Here is the week-one path we use in live codebases. ## Hiring certified engineers - [Claude Implementation Partner Due Diligence: 12 Evidence Checks Before You Hire](https://claudecertified.agency/blog/claude-implementation-partner-due-diligence): A weighted evidence-to-contract scorecard for comparing Claude implementation partners on certified skills, production proof, security, evaluation, ownership, and handoff. - [Claude Engineer Interview Scorecard: Verify Skills Beyond Certification](https://claudecertified.agency/blog/claude-engineer-interview-scorecard): A structured hiring scorecard for verifying Claude engineering capability with credentials, work samples, interview evidence, and production judgment. - [Six weeks to a production Claude agent](https://claudecertified.agency/blog/six-weeks-to-production-claude): Prototype by week two, live traffic by week six. Here is the path we run on a Production Agent Build, and what has to be true on day zero or the calendar is fiction. - [IP ownership when a vendor builds your Claude agent](https://claudecertified.agency/blog/ip-ownership-claude-vendor): If the prompts, evals, and MCP servers do not live in your git from day one, you are renting a demo. Here is the IP rule we sign before week one, and what to reject in a vendor contract. - [Architect vs Developer staffing on a Claude engagement](https://claudecertified.agency/blog/architect-vs-developer-staffing): Developer-certified engineers build the tools and agents. Architect-certified engineers decide what must not be built. Here is how we staff both so you do not get a clever system nobody can govern. - [Audit vs Build vs Embedded: which Claude engagement to buy](https://claudecertified.agency/blog/audit-vs-build-vs-embedded): A two-week audit, a six-week production build, and a quarterly embed solve different problems. Here is how we choose so you do not buy a rebuild when you needed a diagnosis. - [Why Claude certification is a hiring filter, not a badge](https://claudecertified.agency/blog/why-claude-certification-hiring-filter): Most failed Claude projects fail for the same four reasons. Anthropic's exams exist because those reasons are learnable. Here is how we hire and staff against them. ## Study Guide Public CCAR-F study materials. Hierarchy: study guide → curriculum → five exam domains → lessons. Glossary, quick reference, and exercises follow the same five domains. Diagnostic, drill, and the mock exam are practice tools. - [Study guide](https://claudecertified.agency/learn): Hub for the curriculum, exam information, and practice. - [Curriculum](https://claudecertified.agency/learn/curriculum): All 5 domains and 30 lessons, with summaries, key points, and exam focus. ### Domains - [Domain 1: Agentic Architecture & Orchestration](https://claudecertified.agency/learn/1-agentic-architecture): Design and implement agentic systems using Claude's API, including loop management, orchestration patterns, guardrails, and the Claude Agent SDK. - [Domain 2: Tool Design & MCP Integration](https://claudecertified.agency/learn/2-tool-design-mcp): Design effective tool schemas, implement MCP servers and clients, and integrate external services into Claude-powered applications. - [Domain 3: Claude Code Configuration & Workflows](https://claudecertified.agency/learn/3-claude-code-config): Configure Claude Code for development workflows, manage settings, hooks, permissions, and integrate with CI/CD pipelines. - [Domain 4: Prompt Engineering & Structured Output](https://claudecertified.agency/learn/4-prompt-engineering): Craft effective prompts, implement structured output patterns, and apply prompt engineering techniques for production Claude applications. - [Domain 5: Context Management & Reliability](https://claudecertified.agency/learn/5-context-management): Manage context windows effectively, implement caching strategies, handle long conversations, and build reliable production systems. ### Lessons - [1.1 Agentic Loops](https://claudecertified.agency/learn/1-agentic-architecture/1-1-agentic-loops): Lesson in Domain 1, Agentic Architecture & Orchestration. The agentic loop is the core execution cycle: Claude receives a prompt, optionally calls tools, and the loop continues until Claude signals it is finished. Getting the termination condition right is the single most tested concept in this domain. - [1.2 Multi-Agent Orchestration](https://claudecertified.agency/learn/1-agentic-architecture/1-2-orchestration-patterns): Lesson in Domain 1, Agentic Architecture & Orchestration. The exam's multi-agent pattern is hub-and-spoke. A coordinator splits the task, passes context explicitly, and aggregates results. Subagents do not message each other and do not inherit the coordinator's history. - [1.3 Subagent Invocation and Context Passing](https://claudecertified.agency/learn/1-agentic-architecture/1-3-subagent-invocation-context): Lesson in Domain 1, Agentic Architecture & Orchestration. The coordinator spawns subagents with the Task tool (Agent in current Claude Code) and must put Task or Agent in allowedTools. Each subagent receives only the prompt it is given, including structured source metadata. - [1.4 Workflow Enforcement and Handoff](https://claudecertified.agency/learn/1-agentic-architecture/1-4-workflow-enforcement-handoff): Lesson in Domain 1, Agentic Architecture & Orchestration. Task 1.4 is prompt guidance versus a programmatic gate. Money, security, and compliance need code that blocks the tool until its precondition is true. A human handoff has five fields, including the refund amount when one applies. - [1.5 Agent SDK Hooks](https://claudecertified.agency/learn/1-agentic-architecture/1-5-agent-sdk-hooks): Lesson in Domain 1, Agentic Architecture & Orchestration. Agent SDK hooks implement the programmatic side of the 1.4 enforcement spectrum. PreToolUse blocks or rewrites a tool call before it runs. PostToolUse normalises the result after it runs and cannot undo the side effect. - [1.6 Task Decomposition Strategies](https://claudecertified.agency/learn/1-agentic-architecture/1-6-task-decomposition): Lesson in Domain 1, Agentic Architecture & Orchestration. Task decomposition splits complex work into pieces an agent can handle. The exam tests choosing a fixed sequential pipeline or dynamic adaptive decomposition, and recognising attention dilution when too many items share one pass. - [1.7 Session State and Resumption](https://claudecertified.agency/learn/1-agentic-architecture/1-7-session-state-resumption): Lesson in Domain 1, Agentic Architecture & Orchestration. Session management chooses how an agent keeps continuity. --resume continues a valid history. fork_session branches that history to compare approaches. After files change, a fresh session with summary injection is the fix, because a fork still carries the stale tool results. - [2.1 Tool Interface Design](https://claudecertified.agency/learn/2-tool-design-mcp/2-1-tool-schema-design): Lesson in Domain 2, Tool Design & MCP Integration. Tool descriptions are the primary mechanism LLMs use for tool selection. Minimal descriptions such as "Retrieves customer information" cannot separate overlapping tools, so the first fix is to expand those descriptions. - [2.2 Structured Error Responses](https://claudecertified.agency/learn/2-tool-design-mcp/2-2-structured-error-responses): Lesson in Domain 2, Tool Design & MCP Integration. When an MCP tool fails, a generic message such as "Operation failed" gives the agent nothing to recover from. Execution errors return isError: true with structured metadata so the agent can tell a transient failure from a valid empty result. - [2.3 Tool Distribution & Tool Choice](https://claudecertified.agency/learn/2-tool-design-mcp/2-3-tool-distribution-choice): Lesson in Domain 2, Tool Design & MCP Integration. How many tools an agent can see is an architectural decision. Eighteen tools on one agent degrade selection. The workable range is 4-5 tools scoped to the agent's role, with near-duplicates collapsed into one parameterised tool. - [2.4 MCP Server Integration](https://claudecertified.agency/learn/2-tool-design-mcp/2-4-mcp-server-integration): Lesson in Domain 2, Tool Design & MCP Integration. MCP servers connect Claude to databases, APIs, development tools, and issue trackers. Project-level .mcp.json is version-controlled and shared; user-level ~/.claude.json is personal. Credentials use ${ENV_VAR} expansion, community servers come before custom builds, resources catalogue available data, and rich tool descriptions keep MCP tools competitive with built-ins. - [2.5 Built-in Tools](https://claudecertified.agency/learn/2-tool-design-mcp/2-5-built-in-tools): Lesson in Domain 2, Tool Design & MCP Integration. Claude Code's six built-in tools are Read, Write, Edit, Bash, Grep, and Glob. Grep searches file contents; Glob matches file paths. Edit is the default modification. When Edit cannot find a unique anchor, the exam answer is Read + Write, while current Claude Code first widens old_string or uses replace_all. - [3.1 CLAUDE.md Hierarchy, Scoping, and Modular Organisation](https://claudecertified.agency/learn/3-claude-code-config/3-1-claude-md-hierarchy): Lesson in Domain 3, Claude Code Configuration & Workflows. Claude Code loads CLAUDE.md at three levels: user (~/.claude/CLAUDE.md), project (.claude/CLAUDE.md or root CLAUDE.md), and directory. Applicable files are concatenated into context. CLAUDE.md is guidance, not a deterministic enforcement layer — use settings.json or hooks when a rule must hold every run. - [3.2 Custom Slash Commands and Skills](https://claudecertified.agency/learn/3-claude-code-config/3-2-slash-commands-skills): Lesson in Domain 3, Claude Code Configuration & Workflows. Custom commands and skills are one Skills system. .claude/skills//SKILL.md is canonical; .claude/commands/.md still works. A skill is a directory with SKILL.md. A flat .md inside .claude/skills/ does not create a command. Project .claude/ is shared; ~/.claude/ is personal. - [3.3 Path-Specific Rules for Conditional Convention Loading](https://claudecertified.agency/learn/3-claude-code-config/3-3-path-specific-rules): Lesson in Domain 3, Claude Code Configuration & Workflows. Path-specific rules in .claude/rules/ apply conventions only while matching files are being edited. YAML frontmatter paths globs cover a file type spread across many directories. Root CLAUDE.md loads every session, and directory-level CLAUDE.md covers one directory. - [3.4 Plan Mode vs Direct Execution](https://claudecertified.agency/learn/3-claude-code-config/3-4-plan-mode-execution): Lesson in Domain 3, Claude Code Configuration & Workflows. Plan mode explores and designs before any files change. Direct execution applies a known, limited change immediately. The exam choice is ambiguity, not difficulty: architectural and multi-file work is plan mode, often then direct execution, and a clear single-function fix is direct execution. - [3.5 Iterative Refinement Techniques](https://claudecertified.agency/learn/3-claude-code-config/3-5-iterative-refinement): Lesson in Domain 3, Claude Code Configuration & Workflows. Working with Claude Code is iterative. The first output is rarely the final one. The exam checks which technique to reach for first: concrete input/output examples for inconsistent interpretation, test-driven iteration for complex transformations, and the interview pattern for unfamiliar domains. - [3.6 CI/CD Integration](https://claudecertified.agency/learn/3-claude-code-config/3-6-cicd-integration): Lesson in Domain 3, Claude Code Configuration & Workflows. In CI, Claude Code is a non-interactive review and generation engine. The exam tests -p (--print) when a job hangs waiting for input, structured JSON via --output-format json and --json-schema (read structured_output), independent review sessions, incremental findings, and real-time API for blocking pre-merge checks. - [4.1 System Prompts with Explicit Criteria](https://claudecertified.agency/learn/4-prompt-engineering/4-1-system-prompts): Lesson in Domain 4, Prompt Engineering & Structured Output. Vague instructions such as "be conservative" and "only report high-confidence findings" give the model no decision boundary. Explicit categorical criteria define what to flag and what to skip, severity is calibrated with code examples, and a high false-positive category is disabled until its prompt is fixed. - [4.2 Few-Shot Prompting](https://claudecertified.agency/learn/4-prompt-engineering/4-2-few-shot-prompting): Lesson in Domain 4, Prompt Engineering & Structured Output. When detailed instructions still produce inconsistent formatting, ambiguous judgement calls, or empty fields for data that exists, few-shot examples are the first technique. Use 2-4 targeted examples that include reasoning. Malformed JSON, fabricated missing fields, and sum mismatches need other techniques. - [4.3 Structured Output with Tool Use](https://claudecertified.agency/learn/4-prompt-engineering/4-3-structured-output): Lesson in Domain 4, Prompt Engineering & Structured Output. tool_use with a JSON schema eliminates JSON syntax errors. Prompt-based JSON does not. tool_choice auto may return text, any forces some tool call when the document type is unknown, and a named tool forces one step. The schema does not prevent semantic errors, and nullable fields are what stop fabrication. - [4.4 Validation, Retry, and Feedback Loops](https://claudecertified.agency/learn/4-prompt-engineering/4-4-validation-retry-loops): Lesson in Domain 4, Prompt Engineering & Structured Output. Retry-with-error-feedback sends the original document, the failed extraction, and the specific validation error. Retries fix format mismatches, structural errors, misplaced values, and mathematical mistakes. They cannot create information that is absent from the source. tool_use removes schema syntax errors; semantic checks, including Pydantic validators, feed the retry loop. - [4.5 Batch Processing Strategies](https://claudecertified.agency/learn/4-prompt-engineering/4-5-batch-processing): Lesson in Domain 4, Prompt Engineering & Structured Output. The Message Batches API saves 50% against synchronous calls, with a processing window of up to 24 hours and no latency SLA. custom_id matches each request to its result. Blocking work such as a pre-merge check stays synchronous. Overnight debt reports, weekly audits, and nightly test generation use batch. Refine prompts on a sample before you submit, and resubmit only the failed items. - [4.6 Multi-Instance and Multi-Pass Review](https://claudecertified.agency/learn/4-prompt-engineering/4-6-multi-pass-review): Lesson in Domain 4, Prompt Engineering & Structured Output. A model that reviews its own output in the same session keeps the reasoning that produced it and tends to confirm those choices. An independent instance judges the output fresh. Large reviews split into a per-file local pass plus a cross-file integration pass so attention is not diluted. Confidence scores route findings only after labelled sets calibrate the threshold. - [5.1 Context Window Management](https://claudecertified.agency/learn/5-context-management/5-1-context-window-management): Lesson in Domain 5, Context Management & Reliability. Progressive summarisation drops numerical values, dates, percentages, and customer-stated expectations, so a $247.83 refund for order #8891 becomes a recent order. A persistent case facts block, included in every prompt and never summarised, holds those facts. Multi-issue sessions keep a separate structured issue layer. Key findings go at the start of aggregated inputs. Tool results are trimmed before they enter history. The Messages API is stateless, so each request carries the full conversation. Upstream agents return structured findings. Prompt caching exists; implementation details beyond knowing it exists are outside the current exam. - [5.2 Escalation & Ambiguity Resolution](https://claudecertified.agency/learn/5-context-management/5-2-escalation-ambiguity): Lesson in Domain 5, Context Management & Reliability. A support agent escalates for exactly three reasons: the customer explicitly asks for a human, the request is a policy gap or exception, or the agent has tried and cannot make progress. Honour an explicit human request on that turn. A policy violation has a documented refusal; a gap is silent and needs a human. Frustration and a self-reported confidence score are unreliable triggers. A frustrated customer with a straightforward issue gets the resolution; if they then insist on a human, escalate. Multiple customer matches require another identifier. The first fix is explicit escalation criteria with few-shot examples in the system prompt. - [5.3 Error Propagation in Multi-Agent Systems](https://claudecertified.agency/learn/5-context-management/5-3-error-propagation): Lesson in Domain 5, Context Management & Reliability. A failed subagent returns structured error context: failure type, the action it attempted, partial results, and alternative approaches. The four failure types are transient, validation, business, and permission. Silent suppression returns an empty success so the coordinator never recovers. Workflow termination throws away the subagents that finished. An access failure never executed and may be retried. A valid empty result executed and found nothing, so it is the answer. Synthesis output annotates coverage gaps. Transient failures are retried locally before they reach the coordinator. - [5.4 Codebase Exploration & Context Degradation](https://claudecertified.agency/learn/5-context-management/5-4-codebase-exploration): Lesson in Domain 5, Context Management & Reliability. Context degradation is an attention-quality problem: after extended codebase exploration the model cites typical patterns instead of the specific classes, methods, and paths it already found. Verbose discovery output buries earlier findings, and a larger context window does not fix that. Scratchpad files persist those findings outside the conversation and should be maintained from the start. Subagents isolate verbose exploration so the coordinator keeps structured summaries. Phase 1 summaries are injected into Phase 2 prompts to prevent a cold start. /compact is used proactively to protect context quality. A structured state manifest lets a crashed session resume without repeating the exploration. - [5.5 Human Review & Confidence Calibration](https://claudecertified.agency/learn/5-context-management/5-5-human-review-calibration): Lesson in Domain 5, Context Management & Reliability. A 97% aggregate accuracy figure can hide 45–60% accuracy on handwritten receipts, scanned PDFs, and international formats, because standard invoices dominate the volume. Validate accuracy by document type AND field segment before automating. Sample every stratum, including high-confidence extractions that are already automated. Raw confidence is relative: 0.90 on dates can mean 94% actual accuracy and 0.90 on amounts only 82%, so calibrate against labelled validation sets. Route fields above the calibrated threshold to automation with stratified sampling, and put the highest-uncertainty items first in a dynamic reviewer queue. Reduce human review only after that sequence, on segments that stay accurate. - [5.6 Information Provenance & Multi-Source Synthesis](https://claudecertified.agency/learn/5-context-management/5-6-information-provenance): Lesson in Domain 5, Context Management & Reliability. Every finding carries a structured mapping: claim, source URL, document name, relevant excerpt, and publication date. Attribution dies when a synthesis agent paraphrases those mappings away, so each step of the pipeline — research, analysis, synthesis, report — has to merge them forward. When two credible sources disagree, annotate both values with attribution and dates and let the reader decide. Different dates often describe a trend, such as growth moving from 8% to 12%. Render financial data as tables, news as prose, and technical findings as lists. An analysis that hits a conflict finishes with the conflict annotated for the coordinator. ### Glossary - [Glossary](https://claudecertified.agency/learn/glossary): Key terms for each exam domain, with definitions and exam context. - [Glossary — Domain 1: Agentic Architecture & Orchestration](https://claudecertified.agency/learn/glossary/domain-1): Definitions for this domain. - [Glossary — Domain 2: Tool Design & MCP Integration](https://claudecertified.agency/learn/glossary/domain-2): Definitions for this domain. - [Glossary — Domain 3: Claude Code Configuration & Workflows](https://claudecertified.agency/learn/glossary/domain-3): Definitions for this domain. - [Glossary — Domain 4: Prompt Engineering & Structured Output](https://claudecertified.agency/learn/glossary/domain-4): Definitions for this domain. - [Glossary — Domain 5: Context Management & Reliability](https://claudecertified.agency/learn/glossary/domain-5): Definitions for this domain. ### Quick Reference - [Quick reference](https://claudecertified.agency/learn/quick-reference): One-page cheat sheets for each exam domain. - [Quick reference — Domain 1: Agentic Architecture & Orchestration](https://claudecertified.agency/learn/quick-reference/domain-1): Cheat sheet for this domain. - [Quick reference — Domain 2: Tool Design & MCP Integration](https://claudecertified.agency/learn/quick-reference/domain-2): Cheat sheet for this domain. - [Quick reference — Domain 3: Claude Code Configuration & Workflows](https://claudecertified.agency/learn/quick-reference/domain-3): Cheat sheet for this domain. - [Quick reference — Domain 4: Prompt Engineering & Structured Output](https://claudecertified.agency/learn/quick-reference/domain-4): Cheat sheet for this domain. - [Quick reference — Domain 5: Context Management & Reliability](https://claudecertified.agency/learn/quick-reference/domain-5): Cheat sheet for this domain. ### Exercises - [Build exercises](https://claudecertified.agency/learn/exercises): Hands-on projects for every CCAR-F lesson, grouped by domain. - [Exercises — Domain 1: Agentic Architecture & Orchestration](https://claudecertified.agency/learn/exercises/1-agentic-architecture): Build exercises for this domain. - [Exercises — Domain 2: Tool Design & MCP Integration](https://claudecertified.agency/learn/exercises/2-tool-design-mcp): Build exercises for this domain. - [Exercises — Domain 3: Claude Code Configuration & Workflows](https://claudecertified.agency/learn/exercises/3-claude-code-config): Build exercises for this domain. - [Exercises — Domain 4: Prompt Engineering & Structured Output](https://claudecertified.agency/learn/exercises/4-prompt-engineering): Build exercises for this domain. - [Exercises — Domain 5: Context Management & Reliability](https://claudecertified.agency/learn/exercises/5-context-management): Build exercises for this domain. ### Practice - [Diagnostic](https://claudecertified.agency/learn/diagnostic): Timed diagnostic that scores weak areas by domain. - [Drill](https://claudecertified.agency/learn/drill): Short spaced-repetition sessions biased toward missed questions. - [Mock exam](https://claudecertified.agency/mock-exam): Timed mock exams for CCAR-F, CCAR-P, CCAO-F, and CCDV-F with weighted scoring. ## Study Guide (CCAR-P) Public CCAR-P study materials. Hierarchy: study guide → curriculum → seven exam domains → lessons. Glossary, quick reference, and exercises follow the same seven domains. - [Study guide](https://claudecertified.agency/learn/architect-professional): Hub for the Professional curriculum, exam information, and practice. - [Curriculum](https://claudecertified.agency/learn/architect-professional/curriculum): All 7 domains and 38 lessons. ### Domains (Professional) - [Domain 1: Solution Design & Architecture](https://claudecertified.agency/learn/architect-professional/1-solution-design-architecture): Translate business problems into Claude solutions, design end-to-end architectures, choose patterns, and align to value pillars. - [Domain 2: Claude Models, Prompting & Context Engineering](https://claudecertified.agency/learn/architect-professional/2-models-prompting-context): Select models, design prompts and guardrails, engineer context, and reuse prompts efficiently. - [Domain 3: Integration](https://claudecertified.agency/learn/architect-professional/3-integration): Configure tools and agents, close auth gaps, balance latency, observe systems, and design RAG and connection strategies. - [Domain 4: Evaluation, Testing & Optimization](https://claudecertified.agency/learn/architect-professional/4-evaluation-testing-optimization): Define metrics, build evals, run A/B tests, diagnose failures, and optimize cost and performance. - [Domain 5: Governance, Safety & Risk Management](https://claudecertified.agency/learn/architect-professional/5-governance-safety-risk): Implement guardrails, surface risks, apply HITL, meet compliance, and address ethical AI concerns. - [Domain 6: Stakeholder Communication & Lifecycle Management](https://claudecertified.agency/learn/architect-professional/6-stakeholder-lifecycle): Run discovery, communicate trade-offs, manage feedback and SLAs, document designs, and support the full lifecycle. - [Domain 7: Developer Productivity & Operational Enablement](https://claudecertified.agency/learn/architect-professional/7-developer-productivity): Configure Claude for teams, improve AI-assisted workflows, and support debugging and ops. ### Lessons (Professional) - [1.1 Translate business problems into Claude-based AI solutions](https://claudecertified.agency/learn/architect-professional/1-solution-design-architecture/1-1-business-to-claude-solutions): Lesson in Domain 1, Solution Design & Architecture. Map stakeholder goals, constraints, owners, and success metrics to a Claude-shaped scope — decide where language judgment beats deterministic software before picking models, tools, or RAG. - [1.2 Design end-to-end architectures (input → processing → output → feedback loops)](https://claudecertified.agency/learn/architect-professional/1-solution-design-architecture/1-2-end-to-end-architectures): Lesson in Domain 1, Solution Design & Architecture. Design the full path from intake through processing and delivery, plus the feedback loop that captures outcomes, failures, and human review for continuous improvement. - [1.3 Select appropriate architectural patterns (workflow, agentic, augmented LLM)](https://claudecertified.agency/learn/architect-professional/1-solution-design-architecture/1-3-architectural-patterns): Lesson in Domain 1, Solution Design & Architecture. Choose workflow, agentic, or augmented-LLM patterns from task structure, control needs, latency, and unpredictability — not from buzzwords. - [1.4 Design multi-agent systems and orchestration strategies](https://claudecertified.agency/learn/architect-professional/1-solution-design-architecture/1-4-multi-agent-orchestration): Lesson in Domain 1, Solution Design & Architecture. Split roles, pass context explicitly, and orchestrate specialists through a coordinator — no inherited memory, no peer mesh that hides accountability. - [1.5 Apply decomposition techniques for complex problem solving](https://claudecertified.agency/learn/architect-professional/1-solution-design-architecture/1-5-decomposition-techniques): Lesson in Domain 1, Solution Design & Architecture. Break complex work into units that fit model attention, tool scope, and verification — adaptive when structure is unknown, fixed when dependencies are clear. - [1.6 Align solutions to business value pillars (efficiency, transformation, productivity, cost, performance SLAs)](https://claudecertified.agency/learn/architect-professional/1-solution-design-architecture/1-6-business-value-pillars): Lesson in Domain 1, Solution Design & Architecture. Tie every architecture choice to named pillars — efficiency, transformation, productivity, cost, and performance SLAs — with measurable trade-offs stakeholders accepted. - [2.1 Select appropriate Claude models based on trade-offs](https://claudecertified.agency/learn/architect-professional/2-models-prompting-context/2-1-model-selection-tradeoffs): Lesson in Domain 2, Claude Models, Prompting & Context Engineering. Choose relative model tiers by task difficulty and risk, weighing latency and unit cost at expected volume — and revisit when evals or SLAs change. - [2.2 Design system prompts, templates, and guardrails](https://claudecertified.agency/learn/architect-professional/2-models-prompting-context/2-2-system-prompts-templates-guardrails): Lesson in Domain 2, Claude Models, Prompting & Context Engineering. Write system prompts and templates with explicit criteria, formats, and boundaries — and place hard must-always rules on deterministic tool-path controls, not prompts alone. - [2.3 Apply prompt engineering techniques (zero-shot, few-shot, chain-of-thought)](https://claudecertified.agency/learn/architect-professional/2-models-prompting-context/2-3-prompt-engineering-techniques): Lesson in Domain 2, Claude Models, Prompting & Context Engineering. Apply zero-shot, few-shot, and chain-of-thought by failure mode — pick the lightest technique that fixes the problem and earns its tokens under latency SLAs. - [2.4 Optimize context windows and manage token usage](https://claudecertified.agency/learn/architect-professional/2-models-prompting-context/2-4-context-windows-token-usage): Lesson in Domain 2, Claude Models, Prompting & Context Engineering. Budget tokens by layer, keep transactional facts durable, watch position effects, and trim tool results with structured retention — never dump corpora. - [2.5 Implement prompt reuse strategies (caching, modular prompts, Skills)](https://claudecertified.agency/learn/architect-professional/2-models-prompting-context/2-5-prompt-reuse-strategies): Lesson in Domain 2, Claude Models, Prompting & Context Engineering. Reuse prompts via caching, modular shared instructions, and Skills that isolate verbose workflows — keeping live turns lean instead of inlining everything. - [3.1 Evaluate tool/agent configuration for capability bloat](https://claudecertified.agency/learn/architect-professional/3-integration/3-1-tool-agent-capability-bloat): Lesson in Domain 3, Integration. Detect oversized toolkits and agent scopes that degrade tool selection, raise blast radius, and hide unused integrations — then shrink, merge, or split by role before adding more prompt text. - [3.2 Analyze authentication and authorization requirements to identify security gaps](https://claudecertified.agency/learn/architect-professional/3-integration/3-2-auth-security-gaps): Lesson in Domain 3, Integration. Map AuthN vs AuthZ across UI, agent, and tool principals; find gaps where identity succeeds but permission, least privilege, or verified resource binding is missing. - [3.3 Evaluate accuracy-latency trade-offs and justify configuration decisions](https://claudecertified.agency/learn/architect-professional/3-integration/3-3-accuracy-latency-tradeoffs): Lesson in Domain 3, Integration. Justify model, retrieval depth, caching, and tool fan-out against an explicit accuracy floor and latency/cost SLA — quantify deltas and reject one-sided wins that break contracts. - [3.4 Analyze observability challenges and select monitoring strategies at scale](https://claudecertified.agency/learn/architect-professional/3-integration/3-4-observability-at-scale): Lesson in Domain 3, Integration. Instrument LLM-specific SLIs (quality, cost, latency, tool errors), sample traces strategically at high volume, redact secrets, and correlate traces with user outcomes and runbooks. - [3.5 Design a RAG pipeline with appropriate chunking and indexing strategies](https://claudecertified.agency/learn/architect-professional/3-integration/3-5-rag-chunking-indexing): Lesson in Domain 3, Integration. Design structure-aware chunking, provenance metadata, and freshness-safe index promotion so retrieval returns complete, citable units matched to document shape. - [3.6 Apply retrieval strategies matched to data shape and query pattern](https://claudecertified.agency/learn/architect-professional/3-integration/3-6-retrieval-strategies): Lesson in Domain 3, Integration. Classify queries (exact ID, paraphrase, filtered browse) and match dense, sparse, hybrid, and structured filters — evaluate per pattern, not only macro averages. - [3.7 Evaluate connection protocols and select the appropriate integration mechanism (MCP, API/CLI, agent-to-agent)](https://claudecertified.agency/learn/architect-professional/3-integration/3-7-connection-protocols): Lesson in Domain 3, Integration. Choose MCP for shared tools/resources across hosts, API/CLI for owned narrow paths, and agent-to-agent when role isolation and explicit handoffs matter — never secrets-in-prompts. - [3.8 Evaluate progressive discovery vs. monolithic context strategy](https://claudecertified.agency/learn/architect-professional/3-integration/3-8-progressive-vs-monolithic-context): Lesson in Domain 3, Integration. Prefer progressive discovery for large or changing corpora; reserve monolithic (cached) context for small stable packs that fit with headroom — and pin transactional facts outside rolling summaries. - [4.1 Define evaluation metrics (accuracy, latency, cost, safety, security)](https://claudecertified.agency/learn/architect-professional/4-evaluation-testing-optimization/4-1-evaluation-metrics): Lesson in Domain 4, Evaluation, Testing & Optimization. Define a balanced metric set covering accuracy/quality, latency, cost, safety, and security — tied to user and business outcomes, not a single vanity score. - [4.2 Design evaluation datasets and test frameworks using mixed methodologies](https://claudecertified.agency/learn/architect-professional/4-evaluation-testing-optimization/4-2-eval-datasets-frameworks): Lesson in Domain 4, Evaluation, Testing & Optimization. Build versioned eval datasets and harnesses that mix automated checks, human review, and scenario tests — covering happy path, edge, and adversarial cases. - [4.3 Conduct A/B testing and iterative improvements](https://claudecertified.agency/learn/architect-professional/4-evaluation-testing-optimization/4-3-ab-testing-iteration): Lesson in Domain 4, Evaluation, Testing & Optimization. Run controlled A/B tests on prompts, retrieval, and model settings — isolate factors, power the experiment, and ship only when quality and cost clear agreed bars. - [4.4 Diagnose system issues (prompt failure, hallucinations, model mismatch)](https://claudecertified.agency/learn/architect-professional/4-evaluation-testing-optimization/4-4-diagnose-system-issues): Lesson in Domain 4, Evaluation, Testing & Optimization. Separate prompt failure, hallucination/grounding failure, retrieval miss, and model mismatch before fixing — reproduce with fixed inputs and traces. - [4.5 Optimize token usage, latency, and cost-performance trade-offs](https://claudecertified.agency/learn/architect-professional/4-evaluation-testing-optimization/4-5-optimize-token-latency-cost): Lesson in Domain 4, Evaluation, Testing & Optimization. Cut tokens and latency without silently dropping required quality — cache and trim before buying capacity; re-measure quality after each cut; batch when interactivity is not required. - [4.6 Monitor system performance using logging and observability tools](https://claudecertified.agency/learn/architect-professional/4-evaluation-testing-optimization/4-6-monitor-with-observability): Lesson in Domain 4, Evaluation, Testing & Optimization. Operate with logs and observability that catch quality and cost regressions — alert beyond HTTP uptime, keep debug context, and close the loop into eval suites. - [5.1 Implement guardrails and safety controls](https://claudecertified.agency/learn/architect-professional/5-governance-safety-risk/5-1-guardrails-safety-controls): Lesson in Domain 5, Governance, Safety & Risk Management. Layer prompt guidance with deterministic input/output filters, policy checks, and hard blocks on irreversible tool paths — prompts alone never satisfy “must always” safety language. - [5.2 Identify risks, limitations, and failure modes of LLM systems](https://claudecertified.agency/learn/architect-professional/5-governance-safety-risk/5-2-risks-limitations-failures): Lesson in Domain 5, Governance, Safety & Risk Management. Enumerate hallucination, drift, tool misuse, outages, and policy gaps with user impact, mitigations, and residual risk owners — never “the model will handle it.” - [5.3 Apply human-in-the-loop validation strategies](https://claudecertified.agency/learn/architect-professional/5-governance-safety-risk/5-3-human-in-the-loop): Lesson in Domain 5, Governance, Safety & Risk Management. Route high-risk or ambiguous work to humans with measurable escalation triggers and a structured handoff — facts, suggestion, uncertainty, and required decision. - [5.4 Ensure compliance with regulations (e.g., GDPR, HIPAA, FedRAMP)](https://claudecertified.agency/learn/architect-professional/5-governance-safety-risk/5-4-regulatory-compliance): Lesson in Domain 5, Governance, Safety & Risk Management. Map data classes, regions, and flows to the named regulatory regime; minimize sensitive data in prompts and logs; document control ownership and evidence. - [5.5 Address ethical AI considerations (bias, fairness, transparency)](https://claudecertified.agency/learn/architect-professional/5-governance-safety-risk/5-5-ethical-ai-considerations): Lesson in Domain 5, Governance, Safety & Risk Management. Treat bias, fairness, and transparency as design requirements: measure disparate outcomes, disclose AI involvement when required, and provide recourse — not slogans. - [6.1 Conduct structured discovery and requirement gathering](https://claudecertified.agency/learn/architect-professional/6-stakeholder-lifecycle/6-1-structured-discovery): Lesson in Domain 6, Stakeholder Communication & Lifecycle Management. Run structured discovery that surfaces goals, constraints, data, owners, and measurable success metrics — turn findings into testable requirements before picking a stack. - [6.2 Communicate architectural decisions and trade-offs](https://claudecertified.agency/learn/architect-professional/6-stakeholder-lifecycle/6-2-communicate-decisions): Lesson in Domain 6, Stakeholder Communication & Lifecycle Management. Explain decisions with options, trade-offs, costs, risks, and rationale stakeholders can act on — record them in ADRs so rejected alternatives stay visible. - [6.3 Manage stakeholder feedback loops and expectation alignment (including SLAs)](https://claudecertified.agency/learn/architect-professional/6-stakeholder-lifecycle/6-3-stakeholder-feedback-slas): Lesson in Domain 6, Stakeholder Communication & Lifecycle Management. Keep feedback loops and SLA expectations aligned through delivery — publish agreed SLAs, review quality and cost on a cadence, and escalate expectation drift early. - [6.4 Document architectures and provide implementation guidance](https://claudecertified.agency/learn/architect-professional/6-stakeholder-lifecycle/6-4-document-architectures): Lesson in Domain 6, Stakeholder Communication & Lifecycle Management. Document architecture with diagrams, decision rationale, interface contracts, ownership, and failure runbooks so builders and operators can implement safely. - [6.5 Support lifecycle phases (discovery, design, handoff, monitoring, iteration)](https://claudecertified.agency/learn/architect-professional/6-stakeholder-lifecycle/6-5-lifecycle-phases): Lesson in Domain 6, Stakeholder Communication & Lifecycle Management. Support discovery through design, handoff, monitoring, and iteration with explicit exit criteria and ownership — architects who vanish after design miss the lifecycle. - [7.1 Configure Claude tools and environments for teams (e.g., Claude Code)](https://claudecertified.agency/learn/architect-professional/7-developer-productivity/7-1-configure-claude-for-teams): Lesson in Domain 7, Developer Productivity & Operational Enablement. Standardize Claude Code and related tooling with project-scoped shared rules, skills, and MCP — plus secret hygiene and a documented bootstrap — so personal config never stands in for team standards. - [7.2 Improve developer workflows using AI-assisted tooling](https://claudecertified.agency/learn/architect-professional/7-developer-productivity/7-2-ai-assisted-developer-workflows): Lesson in Domain 7, Developer Productivity & Operational Enablement. Embed AI assist into repetitive high-friction steps (review drafts, docs, flaky triage) while humans keep merge/release ownership, and prove value with time saved and error rates — not vanity adoption. - [7.3 Support debugging and operational issue resolution](https://claudecertified.agency/learn/architect-professional/7-developer-productivity/7-3-debugging-operational-issues): Lesson in Domain 7, Developer Productivity & Operational Enablement. Debug Claude, tool, and MCP failures with logs, traces, and minimal fixtures; classify product vs prompt/tool/MCP bugs; capture fixes as reusable runbooks instead of guessing. ### Glossary (Professional) - [Glossary](https://claudecertified.agency/learn/architect-professional/glossary): Key terms for each Professional exam domain. - [Glossary — Domain 1: Solution Design & Architecture](https://claudecertified.agency/learn/architect-professional/glossary/domain-1): Definitions for this domain. - [Glossary — Domain 2: Claude Models, Prompting & Context Engineering](https://claudecertified.agency/learn/architect-professional/glossary/domain-2): Definitions for this domain. - [Glossary — Domain 3: Integration](https://claudecertified.agency/learn/architect-professional/glossary/domain-3): Definitions for this domain. - [Glossary — Domain 4: Evaluation, Testing & Optimization](https://claudecertified.agency/learn/architect-professional/glossary/domain-4): Definitions for this domain. - [Glossary — Domain 5: Governance, Safety & Risk Management](https://claudecertified.agency/learn/architect-professional/glossary/domain-5): Definitions for this domain. - [Glossary — Domain 6: Stakeholder Communication & Lifecycle Management](https://claudecertified.agency/learn/architect-professional/glossary/domain-6): Definitions for this domain. - [Glossary — Domain 7: Developer Productivity & Operational Enablement](https://claudecertified.agency/learn/architect-professional/glossary/domain-7): Definitions for this domain. ### Quick Reference (Professional) - [Quick reference](https://claudecertified.agency/learn/architect-professional/quick-reference): One-page cheat sheets for each Professional domain. - [Quick reference — Domain 1: Solution Design & Architecture](https://claudecertified.agency/learn/architect-professional/quick-reference/domain-1): Cheat sheet for this domain. - [Quick reference — Domain 2: Claude Models, Prompting & Context Engineering](https://claudecertified.agency/learn/architect-professional/quick-reference/domain-2): Cheat sheet for this domain. - [Quick reference — Domain 3: Integration](https://claudecertified.agency/learn/architect-professional/quick-reference/domain-3): Cheat sheet for this domain. - [Quick reference — Domain 4: Evaluation, Testing & Optimization](https://claudecertified.agency/learn/architect-professional/quick-reference/domain-4): Cheat sheet for this domain. - [Quick reference — Domain 5: Governance, Safety & Risk Management](https://claudecertified.agency/learn/architect-professional/quick-reference/domain-5): Cheat sheet for this domain. - [Quick reference — Domain 6: Stakeholder Communication & Lifecycle Management](https://claudecertified.agency/learn/architect-professional/quick-reference/domain-6): Cheat sheet for this domain. - [Quick reference — Domain 7: Developer Productivity & Operational Enablement](https://claudecertified.agency/learn/architect-professional/quick-reference/domain-7): Cheat sheet for this domain. ### Exercises (Professional) - [Build exercises](https://claudecertified.agency/learn/architect-professional/exercises): Hands-on projects for every CCAR-P lesson. - [Exercises — Domain 1: Solution Design & Architecture](https://claudecertified.agency/learn/architect-professional/exercises/1-solution-design-architecture): Exercises for this domain. - [Exercises — Domain 2: Claude Models, Prompting & Context Engineering](https://claudecertified.agency/learn/architect-professional/exercises/2-models-prompting-context): Exercises for this domain. - [Exercises — Domain 3: Integration](https://claudecertified.agency/learn/architect-professional/exercises/3-integration): Exercises for this domain. - [Exercises — Domain 4: Evaluation, Testing & Optimization](https://claudecertified.agency/learn/architect-professional/exercises/4-evaluation-testing-optimization): Exercises for this domain. - [Exercises — Domain 5: Governance, Safety & Risk Management](https://claudecertified.agency/learn/architect-professional/exercises/5-governance-safety-risk): Exercises for this domain. - [Exercises — Domain 6: Stakeholder Communication & Lifecycle Management](https://claudecertified.agency/learn/architect-professional/exercises/6-stakeholder-lifecycle): Exercises for this domain. - [Exercises — Domain 7: Developer Productivity & Operational Enablement](https://claudecertified.agency/learn/architect-professional/exercises/7-developer-productivity): Exercises for this domain. ### Practice (Professional) - [Diagnostic](https://claudecertified.agency/learn/architect-professional/diagnostic): Timed diagnostic scored by Professional domain. - [Drill](https://claudecertified.agency/learn/architect-professional/drill): Short practice sessions biased toward weak areas. - [Mock exam (Professional)](https://claudecertified.agency/mock-exam?track=professional): Timed CCAR-P mock exam (63 questions, weighted). ## Associate Study Guide Public CCAO-F study materials. Hierarchy: study guide → curriculum → seven exam domains → lessons. Glossary, quick reference, and exercises follow the same seven domains. Diagnostic, drill, and the mock exam are practice tools. - [Associate study guide](https://claudecertified.agency/learn/claude-associate): Hub for the CCAO-F curriculum, exam information, and practice. - [Curriculum](https://claudecertified.agency/learn/claude-associate/curriculum): All 7 domains and 30 lessons, with summaries, key points, and exam focus. ### Domains - [Domain 1: Prompting and Task Execution](https://claudecertified.agency/learn/claude-associate/1-prompting-task-execution): Write prompts that give Claude what a capable colleague would need, break big requests into checkable steps, improve results with specific feedback, and tune your approach to the kind of task. - [Domain 2: Output Evaluation and Validation](https://claudecertified.agency/learn/claude-associate/2-output-evaluation-validation): The heaviest domain. Judge whether outputs answer the question, spot hallucinations and bias, verify against a source of record, know when a human must review, and shape outputs for the audience. - [Domain 3: Product and Model Selection](https://claudecertified.agency/learn/claude-associate/3-product-model-selection): Choose the right claude.ai feature for the job, understand the Haiku, Sonnet, and Opus tiers, balance cost, speed, and quality, and work within context limits and memory. - [Domain 4: Workflow Integration and Solution Design](https://claudecertified.agency/learn/claude-associate/4-workflow-integration-design): Turn Claude into part of how work gets done: define the deliverable first, pick strong-fit use cases, pilot and expand, augment or redesign processes, and explain value and limits to stakeholders. - [Domain 5: Configuration and Knowledge Management](https://claudecertified.agency/learn/claude-associate/5-configuration-knowledge): Set up Projects so Claude starts every conversation informed: know which layer holds what, choose between uploads and live connectors, write short standing instructions, and keep configurations current. - [Domain 6: Governance, Risk, and Responsible Use](https://claudecertified.agency/learn/claude-associate/6-governance-risk-responsible-use): Use Claude within the rules: separate what it can do from what you may do, protect sensitive data, follow organisational policy, and take responsibility for ethical outcomes. - [Domain 7: Troubleshooting and Optimization](https://claudecertified.agency/learn/claude-associate/7-troubleshooting-optimization): When results disappoint, diagnose before rewriting: map symptom to cause to first fix, change one variable at a time, and make workflows cheaper and faster without losing quality. ### Lessons - [1.1 Effective Business Prompts](https://claudecertified.agency/learn/claude-associate/1-prompting-task-execution/1-1-effective-business-prompts): A strong business prompt reads like a clear brief to a capable new colleague. It states the role Claude should take, the context behind the request, the constraints that matter, and the format the result should arrive in. Structure beats intensifiers: adding 'be thorough' helps far less than adding the missing context. - [1.2 Task Decomposition](https://claudecertified.agency/learn/claude-associate/1-prompting-task-execution/1-2-task-decomposition): Compound requests fail quietly because Claude has to juggle too many goals in one pass. Splitting the work into a chain of smaller prompts, each with a checkable output, gives you a point to inspect and correct before errors carry forward. - [1.3 Iterative Prompt Improvement](https://claudecertified.agency/learn/claude-associate/1-prompting-task-execution/1-3-iterative-prompt-improvement): The first response is a draft to refine, not a verdict. Improvement comes from specific feedback that says what to keep, what to change, and why. Regenerating and hoping gives you a new random draw instead of progress. - [1.4 Prompt Strategy by Task Type](https://claudecertified.agency/learn/claude-associate/1-prompting-task-execution/1-4-prompt-strategy-by-task-type): Different tasks need different prompting. Analysis wants source material and a clear question. Research wants scope and a way to check claims. Drafting wants audience, voice, and examples. Brainstorming wants room to explore before you narrow. - [2.1 Accuracy and Completeness](https://claudecertified.agency/learn/claude-associate/2-output-evaluation-validation/2-1-accuracy-and-completeness): Before polishing, check two things: did the output answer the question that was asked, and did it leave out anything material? A fluent response that quietly skips a caveat, an exception, or half of the request is still a failed response. - [2.2 Hallucinations, Inconsistencies, and Bias](https://claudecertified.agency/learn/claude-associate/2-output-evaluation-validation/2-2-hallucinations-inconsistencies-bias): Claude can produce text that sounds right but is not grounded in your source or in fact. Learn the red flags: invented numbers and citations, detail beyond what the source contains, internal contradictions, uniform confidence on every claim, and one-sided framing. - [2.3 Fact-Checking and Validation](https://claudecertified.agency/learn/claude-associate/2-output-evaluation-validation/2-3-fact-checking-and-validation): Validation means checking a claim against something independent of the model: the source of record. Self-reported confidence is not validation, and asking Claude to double-check its own answer is weaker than looking at the original document, system, or authority. - [2.4 When Human Review Is Required](https://claudecertified.agency/learn/claude-associate/2-output-evaluation-validation/2-4-when-human-review-is-required): Some outputs should never leave your hands without a person reviewing them. The more external, consequential, or regulated the use, the stronger the need for human review. Claude drafts and assists; accountable people decide. - [2.5 Edit and Adapt for Audience](https://claudecertified.agency/learn/claude-associate/2-output-evaluation-validation/2-5-edit-and-adapt-for-audience): A correct answer can still be wrong for the reader. Adapting tone, length, and level of detail for a specific audience is a normal part of using Claude, and comparing drafts side by side helps you choose deliberately. - [2.6 Output Formats](https://claudecertified.agency/learn/claude-associate/2-output-evaluation-validation/2-6-output-formats): The shape of an answer matters. Short conversational replies suit inline responses, standalone documents and reusable pieces suit Artifacts, and comparisons or lists suit tables and structured layouts. Pick the format that fits how the result will be used. - [3.1 Product Features](https://claudecertified.agency/learn/claude-associate/3-product-model-selection/3-1-product-features): Claude.ai offers several surfaces, and each fits a different kind of work. Chat is for one-off conversation, Projects hold recurring work with shared instructions and knowledge, research mode gathers and synthesises sources, and Artifacts hold standalone deliverables. - [3.2 Model Tiers](https://claudecertified.agency/learn/claude-associate/3-product-model-selection/3-2-model-tiers): Claude comes in tiers with different characters. Haiku is the fast, lightweight option for simple high-volume work. Sonnet is the balanced everyday choice. Opus is the most capable tier for complex reasoning and nuanced judgement. - [3.3 Cost, Speed, and Quality Alignment](https://claudecertified.agency/learn/claude-associate/3-product-model-selection/3-3-cost-speed-quality-alignment): Every task sits somewhere on a triangle of cost, speed, and quality. The skill is picking the smallest model and lightest approach that meets the quality bar, rather than defaulting to the strongest tool everywhere. - [3.4 Context Limits and Memory](https://claudecertified.agency/learn/claude-associate/3-product-model-selection/3-4-context-limits-and-memory): A conversation has a finite context window. Long threads degrade, details get lost, and earlier instructions can be diluted. Know when to restart, when to summarise, and when to persist information in a Project or memory instead of relying on the thread. - [4.1 Requirements and Use Cases](https://claudecertified.agency/learn/claude-associate/4-workflow-integration-design/4-1-requirements-and-use-cases): Good solutions start before the first prompt. Define the deliverable, the audience, the sources, and what success looks like. A clear requirement turns a fuzzy idea like 'use AI for reports' into something Claude and your team can execute and measure. - [4.2 Research, Planning, and Process Optimization](https://claudecertified.agency/learn/claude-associate/4-workflow-integration-design/4-2-research-planning-process-optimization): Claude is strong at some work and weak at other work. Use it for synthesis, drafting, structuring, and analysis of material you provide. Be cautious where the task needs live data, guaranteed precision, or authoritative decisions. - [4.3 Solution Design and Iteration](https://claudecertified.agency/learn/claude-associate/4-workflow-integration-design/4-3-solution-design-and-iteration): Roll out in small steps. Pilot with a narrow use case, measure against the success criteria you defined, learn what breaks, and then expand. Big-bang launches hide problems until they are expensive. - [4.4 Integrate or Redesign Workflows](https://claudecertified.agency/learn/claude-associate/4-workflow-integration-design/4-4-integrate-or-redesign-workflows): You can augment an existing process by dropping Claude into one step, or redesign the process around what Claude makes possible. Augmenting is faster to adopt. Redesigning can unlock bigger gains but needs more change management. - [4.5 Stakeholder Communication](https://claudecertified.agency/learn/claude-associate/4-workflow-integration-design/4-5-stakeholder-communication): When someone asks 'can Claude do this?', give an honest answer that covers value and limits. The best answer describes the division of labour: what Claude does, what people do, and where the review step sits. - [5.1 Projects: Instructions and Knowledge](https://claudecertified.agency/learn/claude-associate/5-configuration-knowledge/5-1-projects-instructions-knowledge): A Project has layers, and each layer has a job. Project instructions hold the standing rules that apply to every conversation. Project knowledge holds reference material. The individual prompt holds what is specific to this one request. - [5.2 Uploaded Knowledge and Connectors](https://claudecertified.agency/learn/claude-associate/5-configuration-knowledge/5-2-uploaded-knowledge-and-connectors): Information can reach Claude by upload or by connector. An uploaded file is a snapshot that goes stale. A connector, such as to a drive or mailbox, reads current content when you use it. Choose based on how often the source changes. - [5.3 System-Level and Project Instructions](https://claudecertified.agency/learn/claude-associate/5-configuration-knowledge/5-3-project-instructions): Standing instructions work best when they are short, specific, and maintained. A tight list of rules Claude can follow beats a long essay of preferences, and instructions need an owner who updates them as the work changes. - [5.4 Maintain and Update Configurations](https://claudecertified.agency/learn/claude-associate/5-configuration-knowledge/5-4-maintain-and-update-configurations): A Project is a living asset. You maintain its knowledge, and stale or cluttered configurations produce confidently wrong results. Replace outdated files, prune what no longer applies, and assign clear ownership for quality. - [6.1 Appropriate vs Inappropriate Use](https://claudecertified.agency/learn/claude-associate/6-governance-risk-responsible-use/6-1-appropriate-vs-inappropriate-use): Capability is not permission. Claude being able to do a task does not mean you are allowed to use it for that task, with that data, in that way. Know the limits of appropriate use and escalate when a need goes beyond the chat product. - [6.2 Data Sensitivity, Privacy, and Regulation](https://claudecertified.agency/learn/claude-associate/6-governance-risk-responsible-use/6-2-data-sensitivity-privacy-regulation): Protect sensitive data before it reaches Claude. The reliable pattern is anonymise first, then analyse. Telling Claude not to retain or reuse information is not a control; removing the sensitive details beforehand is. - [6.3 Organizational AI Policy](https://claudecertified.agency/learn/claude-associate/6-governance-risk-responsible-use/6-3-organizational-ai-policy): Your organisation's AI policy decides what is allowed. Read it before you paste anything, follow it even when it is inconvenient, and escalate when the policy is silent or ambiguous rather than guessing. - [6.4 Ethical Implications](https://claudecertified.agency/learn/claude-associate/6-governance-risk-responsible-use/6-4-ethical-implications): Using AI at work carries ethical duties. Be honest about when AI helped, keep a named person accountable for results, watch for bias, and keep humans in charge of decisions that materially affect people. - [7.1 Diagnose Underperforming Outputs](https://claudecertified.agency/learn/claude-associate/7-troubleshooting-optimization/7-1-diagnose-underperforming-outputs): Before rewriting a prompt, work out what is actually wrong. Move from symptom to cause to first fix. Generic output, wrong facts, ignored instructions, and drift each point to a different cause and a different first remedy. - [7.2 Adjust from Feedback](https://claudecertified.agency/learn/claude-associate/7-troubleshooting-optimization/7-2-adjust-from-feedback): When you change several things at once, you cannot tell which change helped. Adjust one variable at a time, test, and keep what works. Use feedback from reviewers and users as the signal for what to change next. - [7.3 Optimize Workflows](https://claudecertified.agency/learn/claude-associate/7-troubleshooting-optimization/7-3-optimize-workflows): Once a workflow works, make it cheaper and faster without lowering quality. Share what works through Projects, standardise good prompts, and use the smallest model that still meets the bar. - [Glossary](https://claudecertified.agency/learn/claude-associate/glossary): Key terms for each associate exam domain. - [Glossary — Domain 1: Prompting and Task Execution](https://claudecertified.agency/learn/claude-associate/glossary/domain-1): Definitions for this domain. - [Glossary — Domain 2: Output Evaluation and Validation](https://claudecertified.agency/learn/claude-associate/glossary/domain-2): Definitions for this domain. - [Glossary — Domain 3: Product and Model Selection](https://claudecertified.agency/learn/claude-associate/glossary/domain-3): Definitions for this domain. - [Glossary — Domain 4: Workflow Integration and Solution Design](https://claudecertified.agency/learn/claude-associate/glossary/domain-4): Definitions for this domain. - [Glossary — Domain 5: Configuration and Knowledge Management](https://claudecertified.agency/learn/claude-associate/glossary/domain-5): Definitions for this domain. - [Glossary — Domain 6: Governance, Risk, and Responsible Use](https://claudecertified.agency/learn/claude-associate/glossary/domain-6): Definitions for this domain. - [Glossary — Domain 7: Troubleshooting and Optimization](https://claudecertified.agency/learn/claude-associate/glossary/domain-7): Definitions for this domain. - [Quick reference](https://claudecertified.agency/learn/claude-associate/quick-reference): One-page cheat sheets for each associate domain. - [Quick reference — Domain 1](https://claudecertified.agency/learn/claude-associate/quick-reference/domain-1): Cheat sheet for this domain. - [Quick reference — Domain 2](https://claudecertified.agency/learn/claude-associate/quick-reference/domain-2): Cheat sheet for this domain. - [Quick reference — Domain 3](https://claudecertified.agency/learn/claude-associate/quick-reference/domain-3): Cheat sheet for this domain. - [Quick reference — Domain 4](https://claudecertified.agency/learn/claude-associate/quick-reference/domain-4): Cheat sheet for this domain. - [Quick reference — Domain 5](https://claudecertified.agency/learn/claude-associate/quick-reference/domain-5): Cheat sheet for this domain. - [Quick reference — Domain 6](https://claudecertified.agency/learn/claude-associate/quick-reference/domain-6): Cheat sheet for this domain. - [Quick reference — Domain 7](https://claudecertified.agency/learn/claude-associate/quick-reference/domain-7): Cheat sheet for this domain. - [Build exercises](https://claudecertified.agency/learn/claude-associate/exercises): Hands-on projects for every CCAO-F lesson. - [Exercises — Domain 1](https://claudecertified.agency/learn/claude-associate/exercises/1-prompting-task-execution): Exercises for this domain. - [Exercises — Domain 2](https://claudecertified.agency/learn/claude-associate/exercises/2-output-evaluation-validation): Exercises for this domain. - [Exercises — Domain 3](https://claudecertified.agency/learn/claude-associate/exercises/3-product-model-selection): Exercises for this domain. - [Exercises — Domain 4](https://claudecertified.agency/learn/claude-associate/exercises/4-workflow-integration-design): Exercises for this domain. - [Exercises — Domain 5](https://claudecertified.agency/learn/claude-associate/exercises/5-configuration-knowledge): Exercises for this domain. - [Exercises — Domain 6](https://claudecertified.agency/learn/claude-associate/exercises/6-governance-risk-responsible-use): Exercises for this domain. - [Exercises — Domain 7](https://claudecertified.agency/learn/claude-associate/exercises/7-troubleshooting-optimization): Exercises for this domain. - [Diagnostic](https://claudecertified.agency/learn/claude-associate/diagnostic): Timed diagnostic scored by associate domain. - [Drill](https://claudecertified.agency/learn/claude-associate/drill): Short practice sessions biased toward weak areas. ## Developer Study Guide Public CCDV-F materials for all eight exam domains: Agents and Workflows, Applications and Integration, Claude Code, Eval/Testing/Debugging, Model Selection and Optimization, Prompt and Context Engineering, Security and Safety, and Tools and MCPs. Practice questions on these pages are original study items. - [Developer study guide](https://claudecertified.agency/learn/claude-developer): All eight domains for the CCDV-F exam. - [Domain 1: Agents and Workflows](https://claudecertified.agency/learn/claude-developer/domain-1): Decide when a task is a workflow or an agent, construct the loop with the Claude Agent SDK or a harness you own, and apply isolation, memory, and orchestration frameworks only when they earn their cost. - [1.1 Agent Architecture](https://claudecertified.agency/learn/claude-developer/domain-1/agent-architecture): Choose a workflow when the path is known, and an agent when the path is discovered at runtime. A supervisor delegates to isolated subagents. Sequential steps wait on each other. Parallel steps do not. The simplest structure that meets the requirement wins on reliability, predictability, cost, latency, and context. - [1.2 Agent Construction with Claude](https://claudecertified.agency/learn/claude-developer/domain-1/agent-construction): Once an agent is justified, pick who owns the loop. A custom harness gives you the mechanics. The Claude Agent SDK runs the loop, tool dispatch, and sessions in your process. Claude Managed Agents runs the harness for you, on an Anthropic sandbox or a self-hosted one. Hooks and human approval make the steps that must not fail into code. - [1.3 Agent Patterns and Frameworks](https://claudecertified.agency/learn/claude-developer/domain-1/agent-patterns-frameworks): Patterns are how an agent uses tools, memory, and context: a tool loop, isolated subagents, and a window you prune on purpose. Frameworks such as Strands, LangGraph, and PydanticAI standardise loop control, state, and branching when that orchestration is actually complex. They do not make the model more capable. - [Quick reference — Domain 1](https://claudecertified.agency/learn/claude-developer/domain-1/quick-reference): Decision rules and traps for Agents and Workflows. - [Domain 2: Applications and Integration](https://claudecertified.agency/learn/claude-developer/domain-2): Turn a business need into requirements, carry it through the life cycle, and build the Claude application: API mechanics, engineering practice, product design, and configuration. - [2.1 Understanding Requirements](https://claudecertified.agency/learn/claude-developer/domain-2/understanding-requirements): Translate business goals into functional, non-functional, infrastructure, and measurable technical requirements that can drive Claude application architecture. - [2.2 Systems Life Cycle](https://claudecertified.agency/learn/claude-developer/domain-2/systems-life-cycle): A Claude application moves through requirements, design, implementation, verification, release, operation, and change. A model swap or a prompt edit is a new pass through that cycle, not a silent replacement. - [2.3 Claude API Mechanics](https://claudecertified.agency/learn/claude-developer/domain-2/claude-api-mechanics): The Messages API is stateless: the client sends the history every time. Streaming changes when tokens arrive, not what they cost. Batch is for work nobody is waiting on and that does not need a multi-turn tool loop. Prompt caching pays when the stable prefix comes first. - [2.4 Software Engineering Foundations](https://claudecertified.agency/learn/claude-developer/domain-2/software-engineering-foundations): A Claude application is still a service. The Messages call is HTTP and JSON, the client is asynchronous, and prompts, schemas, and model ids belong in version control and in review. Refactor the code into a unit a reviewer can check. - [2.5 Claude Application Design](https://claudecertified.agency/learn/claude-developer/domain-2/claude-application-design): Claude Code, Desktop, claude.ai, the API, and the SDKs do not interpret a product the same way. The application draws a boundary between instructions and untrusted content, declares the schema it will parse, keeps sessions from going stale, and treats plugins as dependencies. - [2.6 Configuration Management](https://claudecertified.agency/learn/claude-developer/domain-2/configuration-management): Shared behavior lives in project configuration. CLAUDE.md is guidance and files are concatenated. settings.json is where a rule is enforced. Model ids, prompts, and plugin versions are pinned so a change is a reviewable diff. - [Quick reference — Domain 2](https://claudecertified.agency/learn/claude-developer/domain-2/quick-reference): Decision rules and traps for Applications and Integration. - [Domain 3: Claude Code](https://claudecertified.agency/learn/claude-developer/domain-3): Operate Claude Code: rules, skills, commands, agents, and memory; sessions and slash commands; headless, streaming, and auto mode; the CLAUDE.md hierarchy; repository initialization; and settings.json. - [3.1 Claude Code Operation](https://claudecertified.agency/learn/claude-developer/domain-3/claude-code-operation): Claude Code is configured with rules, skills, commands, agents, and memory, then operated through sessions, slash commands, and print mode. CLAUDE.md guides the model. settings.json and hooks enforce what the model is allowed to do. - [Quick reference — Domain 3](https://claudecertified.agency/learn/claude-developer/domain-3/quick-reference): Decision rules and traps for Claude Code. - [Domain 4: Eval, Testing, and Debugging](https://claudecertified.agency/learn/claude-developer/domain-4): Identify API errors and stop reasons, choose a recovery that matches the failure, read a trace, and separate an integration-layer bug from model output. - [4.1 Debugging and Error Handling](https://claudecertified.agency/learn/claude-developer/domain-4/debugging-and-error-handling): Name the failure before you change the prompt. An HTTP error never reached a finished answer. A stop_reason tells you whether a successful response is complete, truncated, or waiting on a tool. The trace shows which of those happened, and which layer owns the fix. - [Quick reference — Domain 4](https://claudecertified.agency/learn/claude-developer/domain-4/quick-reference): Decision rules and traps for Eval, Testing, and Debugging. - [Domain 5: Model Selection and Optimization](https://claudecertified.agency/learn/claude-developer/domain-5): Choose how a call is generated, how it reaches the API, which family serves it, and what the tokens cost: fundamentals, the REST client, Opus versus Sonnet versus Haiku, and caching. - [5.1 LLM Fundamentals](https://claudecertified.agency/learn/claude-developer/domain-5/llm-fundamentals): Claude generates one next token at a time inside a finite context window. Sampling makes that choice vary. Thinking, effort, fast mode, and examples change how many tokens that process spends and what shape the text takes. - [5.2 Technical Fundamentals](https://claudecertified.agency/learn/claude-developer/domain-5/technical-fundamentals): The Claude API is HTTP. Official SDKs wrap that REST surface with types, retries, and streaming helpers. Token streaming is server-sent events on that HTTP response. A WebSocket is a different, two-way connection, and it is not how the Messages API delivers tokens. - [5.3 Model Selection and Tradeoffs](https://claudecertified.agency/learn/claude-developer/domain-5/model-selection-and-tradeoffs): Haiku, Sonnet, and Opus are a speed, quality, and price ladder. Effort and fast mode move a call along that ladder without a new name. A new model id is a behavior change: pin the snapshot you evaluated, and rerun the eval before you serve it. - [5.4 Cost and Token Management](https://claudecertified.agency/learn/claude-developer/domain-5/cost-and-token-management): The bill is the usage object times the price card. Output and uncached input dominate. Prompt caching stores a shared prefix at a cache checkpoint so later calls read it at a fraction of the input price. The checkpoint only hits when that prefix, including thinking and effort, stays the same. - [Quick reference — Domain 5](https://claudecertified.agency/learn/claude-developer/domain-5/quick-reference): Decision rules and traps for Model Selection and Optimization. - [Domain 6: Prompt and Context Engineering](https://claudecertified.agency/learn/claude-developer/domain-6): Decide what stays in the window, how the instruction and the case are placed, and how a program checks the text that comes back. - [6.1 Context Engineering](https://claudecertified.agency/learn/claude-developer/domain-6/context-engineering): The context window is the only memory a call has. Drift and bloat happen when old tool dumps crowd out the instruction. Prune tool output, compact a long transcript into a summary, and isolate a subtask so its scratch work never enters the parent. - [6.2 Prompt Engineering](https://claudecertified.agency/learn/claude-developer/domain-6/prompt-engineering): A prompt is a placement problem as much as a wording problem. Durable rules go in the system turn, this call's data goes in the user turn, and examples cover the edge the rules keep missing. Change one thing, score it, and keep untrusted text in a data slot. - [6.3 Output Handling](https://claudecertified.agency/learn/claude-developer/domain-6/output-handling): A confident answer is still model output. Structured output makes the shape parseable. Your code validates the values, parses defensively, and refuses a field the source does not support. - [Quick reference — Domain 6](https://claudecertified.agency/learn/claude-developer/domain-6/quick-reference): Decision rules and traps for Prompt and Context Engineering. - [Domain 7: Security and Safety](https://claudecertified.agency/learn/claude-developer/domain-7): Build Claude applications with secure data handling, layered guardrails, safe tool execution, and controlled identity, secrets, and access management. - [7.1 AI Application Security](https://claudecertified.agency/learn/claude-developer/domain-7/ai-application-security): Protect Claude applications from prompt injection, jailbreaks, untrusted input, data leakage, and unauthorized access while preserving confidentiality, privacy, and integrity. - [7.2 Guardrails and Safe Deployment](https://claudecertified.agency/learn/claude-developer/domain-7/guardrails-safe-deployment): Apply layered guardrails and secure-by-design practices to deploy Claude applications safely, with privacy, identity, access control, and least privilege. - [7.3 Claude Hooks](https://claudecertified.agency/learn/claude-developer/domain-7/claude-hooks): Use Claude hooks to enforce guardrails, validate tool activity, and prevent destructive or unauthorized actions in Claude applications. - [7.4 Identity, Secrets, and Key Management](https://claudecertified.agency/learn/claude-developer/domain-7/identity-secrets-key-management): Manage identities, credentials, secrets, and API keys securely across Claude development and production environments. - [Quick reference — Domain 7](https://claudecertified.agency/learn/claude-developer/domain-7/quick-reference): Decision rules and traps for Security and Safety. - [Domain 8: Tools and MCPs](https://claudecertified.agency/learn/claude-developer/domain-8): Design reliable Claude tool integrations, build MCP servers, and choose appropriately among built-in tools, custom tools, Skills, and MCPs. - [8.1 Tool Implementation](https://claudecertified.agency/learn/claude-developer/domain-8/tool-implementation): Design and implement Claude tools for reliable external-system interaction, including clear tool descriptions, constrained inputs, error handling, dispatch patterns, approval flows, and practical tool-set construction. - [8.2 MCP Server Development](https://claudecertified.agency/learn/claude-developer/domain-8/mcp-server-development): Build and integrate MCP servers that expose tools, resources, and prompts to Claude applications while choosing appropriate deployment and communication patterns. - [8.3 Agentic Customization](https://claudecertified.agency/learn/claude-developer/domain-8/agentic-customization): Choose the appropriate agentic customization mechanism by evaluating the tradeoffs among built-in tools, custom tools, Skills, and MCPs. - [Quick reference — Domain 8](https://claudecertified.agency/learn/claude-developer/domain-8/quick-reference): Decision rules and traps for Tools and MCPs. - [Quick reference](https://claudecertified.agency/learn/claude-developer/quick-reference): Cheat sheets for all published domains. - [Build exercises](https://claudecertified.agency/learn/claude-developer/exercises): Hands-on projects from the published topics. - [Diagnostic](https://claudecertified.agency/learn/claude-developer/diagnostic): Timed diagnostic scored from the published domains. - [Glossary](https://claudecertified.agency/learn/claude-developer/glossary): Terms defined on the published topic pages. - [Drill](https://claudecertified.agency/learn/claude-developer/drill): Short sessions from the published topic questions, biased toward misses. ## Optional - [llms-full.txt](https://claudecertified.agency/llms-full.txt): Complete URL index including every study path. - [RSS](https://claudecertified.agency/blog/rss.xml): Full post index as RSS 2.0. - [Sitemap](https://claudecertified.agency/sitemap.xml): Canonical URL list for crawlers. ## Complete study URL index Every public study and marketing URL from the sitemap (progress dashboards omitted; they are noindex). - https://claudecertified.agency/certifications - https://claudecertified.agency/faq - https://claudecertified.agency/services - https://claudecertified.agency/process - https://claudecertified.agency/bench - https://claudecertified.agency/learn - https://claudecertified.agency/learn/curriculum - https://claudecertified.agency/learn/1-agentic-architecture - https://claudecertified.agency/learn/2-tool-design-mcp - https://claudecertified.agency/learn/3-claude-code-config - https://claudecertified.agency/learn/4-prompt-engineering - https://claudecertified.agency/learn/5-context-management - https://claudecertified.agency/learn/1-agentic-architecture/1-1-agentic-loops - https://claudecertified.agency/learn/1-agentic-architecture/1-2-orchestration-patterns - https://claudecertified.agency/learn/1-agentic-architecture/1-3-subagent-invocation-context - https://claudecertified.agency/learn/1-agentic-architecture/1-4-workflow-enforcement-handoff - https://claudecertified.agency/learn/1-agentic-architecture/1-5-agent-sdk-hooks - https://claudecertified.agency/learn/1-agentic-architecture/1-6-task-decomposition - https://claudecertified.agency/learn/1-agentic-architecture/1-7-session-state-resumption - https://claudecertified.agency/learn/2-tool-design-mcp/2-1-tool-schema-design - https://claudecertified.agency/learn/2-tool-design-mcp/2-2-structured-error-responses - https://claudecertified.agency/learn/2-tool-design-mcp/2-3-tool-distribution-choice - https://claudecertified.agency/learn/2-tool-design-mcp/2-4-mcp-server-integration - https://claudecertified.agency/learn/2-tool-design-mcp/2-5-built-in-tools - https://claudecertified.agency/learn/3-claude-code-config/3-1-claude-md-hierarchy - https://claudecertified.agency/learn/3-claude-code-config/3-2-slash-commands-skills - https://claudecertified.agency/learn/3-claude-code-config/3-3-path-specific-rules - https://claudecertified.agency/learn/3-claude-code-config/3-4-plan-mode-execution - https://claudecertified.agency/learn/3-claude-code-config/3-5-iterative-refinement - https://claudecertified.agency/learn/3-claude-code-config/3-6-cicd-integration - https://claudecertified.agency/learn/4-prompt-engineering/4-1-system-prompts - https://claudecertified.agency/learn/4-prompt-engineering/4-2-few-shot-prompting - https://claudecertified.agency/learn/4-prompt-engineering/4-3-structured-output - https://claudecertified.agency/learn/4-prompt-engineering/4-4-validation-retry-loops - https://claudecertified.agency/learn/4-prompt-engineering/4-5-batch-processing - https://claudecertified.agency/learn/4-prompt-engineering/4-6-multi-pass-review - https://claudecertified.agency/learn/5-context-management/5-1-context-window-management - https://claudecertified.agency/learn/5-context-management/5-2-escalation-ambiguity - https://claudecertified.agency/learn/5-context-management/5-3-error-propagation - https://claudecertified.agency/learn/5-context-management/5-4-codebase-exploration - https://claudecertified.agency/learn/5-context-management/5-5-human-review-calibration - https://claudecertified.agency/learn/5-context-management/5-6-information-provenance - https://claudecertified.agency/learn/glossary - https://claudecertified.agency/learn/glossary/domain-1 - https://claudecertified.agency/learn/glossary/domain-2 - https://claudecertified.agency/learn/glossary/domain-3 - https://claudecertified.agency/learn/glossary/domain-4 - https://claudecertified.agency/learn/glossary/domain-5 - https://claudecertified.agency/learn/quick-reference - https://claudecertified.agency/learn/quick-reference/domain-1 - https://claudecertified.agency/learn/quick-reference/domain-2 - https://claudecertified.agency/learn/quick-reference/domain-3 - https://claudecertified.agency/learn/quick-reference/domain-4 - https://claudecertified.agency/learn/quick-reference/domain-5 - https://claudecertified.agency/learn/exercises - https://claudecertified.agency/learn/exercises/1-agentic-architecture - https://claudecertified.agency/learn/exercises/2-tool-design-mcp - https://claudecertified.agency/learn/exercises/3-claude-code-config - https://claudecertified.agency/learn/exercises/4-prompt-engineering - https://claudecertified.agency/learn/exercises/5-context-management - https://claudecertified.agency/learn/diagnostic - https://claudecertified.agency/learn/drill - https://claudecertified.agency/learn/claude-associate - https://claudecertified.agency/learn/claude-associate/curriculum - https://claudecertified.agency/learn/claude-associate/1-prompting-task-execution - https://claudecertified.agency/learn/claude-associate/2-output-evaluation-validation - https://claudecertified.agency/learn/claude-associate/3-product-model-selection - https://claudecertified.agency/learn/claude-associate/4-workflow-integration-design - https://claudecertified.agency/learn/claude-associate/5-configuration-knowledge - https://claudecertified.agency/learn/claude-associate/6-governance-risk-responsible-use - https://claudecertified.agency/learn/claude-associate/7-troubleshooting-optimization - https://claudecertified.agency/learn/claude-associate/1-prompting-task-execution/1-1-effective-business-prompts - https://claudecertified.agency/learn/claude-associate/1-prompting-task-execution/1-2-task-decomposition - https://claudecertified.agency/learn/claude-associate/1-prompting-task-execution/1-3-iterative-prompt-improvement - https://claudecertified.agency/learn/claude-associate/1-prompting-task-execution/1-4-prompt-strategy-by-task-type - https://claudecertified.agency/learn/claude-associate/2-output-evaluation-validation/2-1-accuracy-and-completeness - https://claudecertified.agency/learn/claude-associate/2-output-evaluation-validation/2-2-hallucinations-inconsistencies-bias - https://claudecertified.agency/learn/claude-associate/2-output-evaluation-validation/2-3-fact-checking-and-validation - https://claudecertified.agency/learn/claude-associate/2-output-evaluation-validation/2-4-when-human-review-is-required - https://claudecertified.agency/learn/claude-associate/2-output-evaluation-validation/2-5-edit-and-adapt-for-audience - https://claudecertified.agency/learn/claude-associate/2-output-evaluation-validation/2-6-output-formats - https://claudecertified.agency/learn/claude-associate/3-product-model-selection/3-1-product-features - https://claudecertified.agency/learn/claude-associate/3-product-model-selection/3-2-model-tiers - https://claudecertified.agency/learn/claude-associate/3-product-model-selection/3-3-cost-speed-quality-alignment - https://claudecertified.agency/learn/claude-associate/3-product-model-selection/3-4-context-limits-and-memory - https://claudecertified.agency/learn/claude-associate/4-workflow-integration-design/4-1-requirements-and-use-cases - https://claudecertified.agency/learn/claude-associate/4-workflow-integration-design/4-2-research-planning-process-optimization - https://claudecertified.agency/learn/claude-associate/4-workflow-integration-design/4-3-solution-design-and-iteration - https://claudecertified.agency/learn/claude-associate/4-workflow-integration-design/4-4-integrate-or-redesign-workflows - https://claudecertified.agency/learn/claude-associate/4-workflow-integration-design/4-5-stakeholder-communication - https://claudecertified.agency/learn/claude-associate/5-configuration-knowledge/5-1-projects-instructions-knowledge - https://claudecertified.agency/learn/claude-associate/5-configuration-knowledge/5-2-uploaded-knowledge-and-connectors - https://claudecertified.agency/learn/claude-associate/5-configuration-knowledge/5-3-project-instructions - https://claudecertified.agency/learn/claude-associate/5-configuration-knowledge/5-4-maintain-and-update-configurations - https://claudecertified.agency/learn/claude-associate/6-governance-risk-responsible-use/6-1-appropriate-vs-inappropriate-use - https://claudecertified.agency/learn/claude-associate/6-governance-risk-responsible-use/6-2-data-sensitivity-privacy-regulation - https://claudecertified.agency/learn/claude-associate/6-governance-risk-responsible-use/6-3-organizational-ai-policy - https://claudecertified.agency/learn/claude-associate/6-governance-risk-responsible-use/6-4-ethical-implications - https://claudecertified.agency/learn/claude-associate/7-troubleshooting-optimization/7-1-diagnose-underperforming-outputs - https://claudecertified.agency/learn/claude-associate/7-troubleshooting-optimization/7-2-adjust-from-feedback - https://claudecertified.agency/learn/claude-associate/7-troubleshooting-optimization/7-3-optimize-workflows - https://claudecertified.agency/learn/claude-associate/glossary - https://claudecertified.agency/learn/claude-associate/glossary/domain-1 - https://claudecertified.agency/learn/claude-associate/glossary/domain-2 - https://claudecertified.agency/learn/claude-associate/glossary/domain-3 - https://claudecertified.agency/learn/claude-associate/glossary/domain-4 - https://claudecertified.agency/learn/claude-associate/glossary/domain-5 - https://claudecertified.agency/learn/claude-associate/glossary/domain-6 - https://claudecertified.agency/learn/claude-associate/glossary/domain-7 - https://claudecertified.agency/learn/claude-associate/quick-reference - https://claudecertified.agency/learn/claude-associate/quick-reference/domain-1 - https://claudecertified.agency/learn/claude-associate/quick-reference/domain-2 - https://claudecertified.agency/learn/claude-associate/quick-reference/domain-3 - https://claudecertified.agency/learn/claude-associate/quick-reference/domain-4 - https://claudecertified.agency/learn/claude-associate/quick-reference/domain-5 - https://claudecertified.agency/learn/claude-associate/quick-reference/domain-6 - https://claudecertified.agency/learn/claude-associate/quick-reference/domain-7 - https://claudecertified.agency/learn/claude-associate/exercises - https://claudecertified.agency/learn/claude-associate/exercises/1-prompting-task-execution - https://claudecertified.agency/learn/claude-associate/exercises/2-output-evaluation-validation - https://claudecertified.agency/learn/claude-associate/exercises/3-product-model-selection - https://claudecertified.agency/learn/claude-associate/exercises/4-workflow-integration-design - https://claudecertified.agency/learn/claude-associate/exercises/5-configuration-knowledge - https://claudecertified.agency/learn/claude-associate/exercises/6-governance-risk-responsible-use - https://claudecertified.agency/learn/claude-associate/exercises/7-troubleshooting-optimization - https://claudecertified.agency/learn/claude-associate/diagnostic - https://claudecertified.agency/learn/claude-associate/drill - https://claudecertified.agency/learn/claude-developer - https://claudecertified.agency/learn/claude-developer/domain-1 - https://claudecertified.agency/learn/claude-developer/domain-1/agent-architecture - https://claudecertified.agency/learn/claude-developer/domain-1/agent-construction - https://claudecertified.agency/learn/claude-developer/domain-1/agent-patterns-frameworks - https://claudecertified.agency/learn/claude-developer/domain-1/quick-reference - https://claudecertified.agency/learn/claude-developer/domain-2 - https://claudecertified.agency/learn/claude-developer/domain-2/understanding-requirements - https://claudecertified.agency/learn/claude-developer/domain-2/systems-life-cycle - https://claudecertified.agency/learn/claude-developer/domain-2/claude-api-mechanics - https://claudecertified.agency/learn/claude-developer/domain-2/software-engineering-foundations - https://claudecertified.agency/learn/claude-developer/domain-2/claude-application-design - https://claudecertified.agency/learn/claude-developer/domain-2/configuration-management - https://claudecertified.agency/learn/claude-developer/domain-2/quick-reference - https://claudecertified.agency/learn/claude-developer/domain-3 - https://claudecertified.agency/learn/claude-developer/domain-3/claude-code-operation - https://claudecertified.agency/learn/claude-developer/domain-3/quick-reference - https://claudecertified.agency/learn/claude-developer/domain-4 - https://claudecertified.agency/learn/claude-developer/domain-4/debugging-and-error-handling - https://claudecertified.agency/learn/claude-developer/domain-4/quick-reference - https://claudecertified.agency/learn/claude-developer/domain-5 - https://claudecertified.agency/learn/claude-developer/domain-5/llm-fundamentals - https://claudecertified.agency/learn/claude-developer/domain-5/technical-fundamentals - https://claudecertified.agency/learn/claude-developer/domain-5/model-selection-and-tradeoffs - https://claudecertified.agency/learn/claude-developer/domain-5/cost-and-token-management - https://claudecertified.agency/learn/claude-developer/domain-5/quick-reference - https://claudecertified.agency/learn/claude-developer/domain-6 - https://claudecertified.agency/learn/claude-developer/domain-6/context-engineering - https://claudecertified.agency/learn/claude-developer/domain-6/prompt-engineering - https://claudecertified.agency/learn/claude-developer/domain-6/output-handling - https://claudecertified.agency/learn/claude-developer/domain-6/quick-reference - https://claudecertified.agency/learn/claude-developer/domain-7 - https://claudecertified.agency/learn/claude-developer/domain-7/ai-application-security - https://claudecertified.agency/learn/claude-developer/domain-7/guardrails-safe-deployment - https://claudecertified.agency/learn/claude-developer/domain-7/claude-hooks - https://claudecertified.agency/learn/claude-developer/domain-7/identity-secrets-key-management - https://claudecertified.agency/learn/claude-developer/domain-7/quick-reference - https://claudecertified.agency/learn/claude-developer/domain-8 - https://claudecertified.agency/learn/claude-developer/domain-8/tool-implementation - https://claudecertified.agency/learn/claude-developer/domain-8/mcp-server-development - https://claudecertified.agency/learn/claude-developer/domain-8/agentic-customization - https://claudecertified.agency/learn/claude-developer/domain-8/quick-reference - https://claudecertified.agency/learn/claude-developer/quick-reference - https://claudecertified.agency/learn/claude-developer/exercises - https://claudecertified.agency/learn/claude-developer/diagnostic - https://claudecertified.agency/learn/claude-developer/glossary - https://claudecertified.agency/learn/claude-developer/drill - https://claudecertified.agency/learn/claude-developer/quick-reference/domain-1 - https://claudecertified.agency/learn/claude-developer/quick-reference/domain-2 - https://claudecertified.agency/learn/claude-developer/quick-reference/domain-3 - https://claudecertified.agency/learn/claude-developer/quick-reference/domain-4 - https://claudecertified.agency/learn/claude-developer/quick-reference/domain-5 - https://claudecertified.agency/learn/claude-developer/quick-reference/domain-6 - https://claudecertified.agency/learn/claude-developer/quick-reference/domain-7 - https://claudecertified.agency/learn/claude-developer/quick-reference/domain-8 - https://claudecertified.agency/learn/claude-developer/exercises/domain-1 - https://claudecertified.agency/learn/claude-developer/exercises/domain-2 - https://claudecertified.agency/learn/claude-developer/exercises/domain-3 - https://claudecertified.agency/learn/claude-developer/exercises/domain-4 - https://claudecertified.agency/learn/claude-developer/exercises/domain-5 - https://claudecertified.agency/learn/claude-developer/exercises/domain-6 - https://claudecertified.agency/learn/claude-developer/exercises/domain-7 - https://claudecertified.agency/learn/claude-developer/exercises/domain-8 - https://claudecertified.agency/learn/claude-developer/glossary/domain-1 - https://claudecertified.agency/learn/claude-developer/glossary/domain-2 - https://claudecertified.agency/learn/claude-developer/glossary/domain-3 - https://claudecertified.agency/learn/claude-developer/glossary/domain-4 - https://claudecertified.agency/learn/claude-developer/glossary/domain-5 - https://claudecertified.agency/learn/claude-developer/glossary/domain-6 - https://claudecertified.agency/learn/claude-developer/glossary/domain-7 - https://claudecertified.agency/learn/claude-developer/glossary/domain-8 - https://claudecertified.agency/mock-exam - https://claudecertified.agency/learn/architect-professional - https://claudecertified.agency/learn/architect-professional/curriculum - https://claudecertified.agency/learn/architect-professional/1-solution-design-architecture - https://claudecertified.agency/learn/architect-professional/2-models-prompting-context - https://claudecertified.agency/learn/architect-professional/3-integration - https://claudecertified.agency/learn/architect-professional/4-evaluation-testing-optimization - https://claudecertified.agency/learn/architect-professional/5-governance-safety-risk - https://claudecertified.agency/learn/architect-professional/6-stakeholder-lifecycle - https://claudecertified.agency/learn/architect-professional/7-developer-productivity - https://claudecertified.agency/learn/architect-professional/1-solution-design-architecture/1-1-business-to-claude-solutions - https://claudecertified.agency/learn/architect-professional/1-solution-design-architecture/1-2-end-to-end-architectures - https://claudecertified.agency/learn/architect-professional/1-solution-design-architecture/1-3-architectural-patterns - https://claudecertified.agency/learn/architect-professional/1-solution-design-architecture/1-4-multi-agent-orchestration - https://claudecertified.agency/learn/architect-professional/1-solution-design-architecture/1-5-decomposition-techniques - https://claudecertified.agency/learn/architect-professional/1-solution-design-architecture/1-6-business-value-pillars - https://claudecertified.agency/learn/architect-professional/2-models-prompting-context/2-1-model-selection-tradeoffs - https://claudecertified.agency/learn/architect-professional/2-models-prompting-context/2-2-system-prompts-templates-guardrails - https://claudecertified.agency/learn/architect-professional/2-models-prompting-context/2-3-prompt-engineering-techniques - https://claudecertified.agency/learn/architect-professional/2-models-prompting-context/2-4-context-windows-token-usage - https://claudecertified.agency/learn/architect-professional/2-models-prompting-context/2-5-prompt-reuse-strategies - https://claudecertified.agency/learn/architect-professional/3-integration/3-1-tool-agent-capability-bloat - https://claudecertified.agency/learn/architect-professional/3-integration/3-2-auth-security-gaps - https://claudecertified.agency/learn/architect-professional/3-integration/3-3-accuracy-latency-tradeoffs - https://claudecertified.agency/learn/architect-professional/3-integration/3-4-observability-at-scale - https://claudecertified.agency/learn/architect-professional/3-integration/3-5-rag-chunking-indexing - https://claudecertified.agency/learn/architect-professional/3-integration/3-6-retrieval-strategies - https://claudecertified.agency/learn/architect-professional/3-integration/3-7-connection-protocols - https://claudecertified.agency/learn/architect-professional/3-integration/3-8-progressive-vs-monolithic-context - https://claudecertified.agency/learn/architect-professional/4-evaluation-testing-optimization/4-1-evaluation-metrics - https://claudecertified.agency/learn/architect-professional/4-evaluation-testing-optimization/4-2-eval-datasets-frameworks - https://claudecertified.agency/learn/architect-professional/4-evaluation-testing-optimization/4-3-ab-testing-iteration - https://claudecertified.agency/learn/architect-professional/4-evaluation-testing-optimization/4-4-diagnose-system-issues - https://claudecertified.agency/learn/architect-professional/4-evaluation-testing-optimization/4-5-optimize-token-latency-cost - https://claudecertified.agency/learn/architect-professional/4-evaluation-testing-optimization/4-6-monitor-with-observability - https://claudecertified.agency/learn/architect-professional/5-governance-safety-risk/5-1-guardrails-safety-controls - https://claudecertified.agency/learn/architect-professional/5-governance-safety-risk/5-2-risks-limitations-failures - https://claudecertified.agency/learn/architect-professional/5-governance-safety-risk/5-3-human-in-the-loop - https://claudecertified.agency/learn/architect-professional/5-governance-safety-risk/5-4-regulatory-compliance - https://claudecertified.agency/learn/architect-professional/5-governance-safety-risk/5-5-ethical-ai-considerations - https://claudecertified.agency/learn/architect-professional/6-stakeholder-lifecycle/6-1-structured-discovery - https://claudecertified.agency/learn/architect-professional/6-stakeholder-lifecycle/6-2-communicate-decisions - https://claudecertified.agency/learn/architect-professional/6-stakeholder-lifecycle/6-3-stakeholder-feedback-slas - https://claudecertified.agency/learn/architect-professional/6-stakeholder-lifecycle/6-4-document-architectures - https://claudecertified.agency/learn/architect-professional/6-stakeholder-lifecycle/6-5-lifecycle-phases - https://claudecertified.agency/learn/architect-professional/7-developer-productivity/7-1-configure-claude-for-teams - https://claudecertified.agency/learn/architect-professional/7-developer-productivity/7-2-ai-assisted-developer-workflows - https://claudecertified.agency/learn/architect-professional/7-developer-productivity/7-3-debugging-operational-issues - https://claudecertified.agency/learn/architect-professional/glossary - https://claudecertified.agency/learn/architect-professional/glossary/domain-1 - https://claudecertified.agency/learn/architect-professional/glossary/domain-2 - https://claudecertified.agency/learn/architect-professional/glossary/domain-3 - https://claudecertified.agency/learn/architect-professional/glossary/domain-4 - https://claudecertified.agency/learn/architect-professional/glossary/domain-5 - https://claudecertified.agency/learn/architect-professional/glossary/domain-6 - https://claudecertified.agency/learn/architect-professional/glossary/domain-7 - https://claudecertified.agency/learn/architect-professional/quick-reference - https://claudecertified.agency/learn/architect-professional/quick-reference/domain-1 - https://claudecertified.agency/learn/architect-professional/quick-reference/domain-2 - https://claudecertified.agency/learn/architect-professional/quick-reference/domain-3 - https://claudecertified.agency/learn/architect-professional/quick-reference/domain-4 - https://claudecertified.agency/learn/architect-professional/quick-reference/domain-5 - https://claudecertified.agency/learn/architect-professional/quick-reference/domain-6 - https://claudecertified.agency/learn/architect-professional/quick-reference/domain-7 - https://claudecertified.agency/learn/architect-professional/exercises - https://claudecertified.agency/learn/architect-professional/exercises/1-solution-design-architecture - https://claudecertified.agency/learn/architect-professional/exercises/2-models-prompting-context - https://claudecertified.agency/learn/architect-professional/exercises/3-integration - https://claudecertified.agency/learn/architect-professional/exercises/4-evaluation-testing-optimization - https://claudecertified.agency/learn/architect-professional/exercises/5-governance-safety-risk - https://claudecertified.agency/learn/architect-professional/exercises/6-stakeholder-lifecycle - https://claudecertified.agency/learn/architect-professional/exercises/7-developer-productivity - https://claudecertified.agency/learn/architect-professional/diagnostic - https://claudecertified.agency/learn/architect-professional/drill