This is the single starting point for your AIP-C01 prep. The flow below is designed for speed: start with the Study Guide to build the map, reinforce with Flashcards, pressure-test with the Practice Exam and Quick Drill, then finish with Final Practice for last-mile work.
This order is optimized for the next 2 days: learn, reinforce, pressure-test, then correct.
Start here to get the exam blueprint, domain structure, service map, decision trees, and the highest-yield exam traps in one place.
Move here for flashcards, rapid drills, pattern matching, readiness tracking, and the new 48-hour focus board.
Use this when you want a broader exam-style run with more volume, domain filters, flags, and review mode.
Good for quick exam-style runs and mistake review from the extracted question bank.
Finish here. This is your last-mile pressure deck focused on AWS-native distinctions and common trap answers.
Agentic AI architecture, model deployment strategies, cost & performance optimization, and caching patterns. Covers the most under-studied 40% of the exam.
CloudWatch monitoring, X-Ray tracing, RAG troubleshooting framework, RAGAS evaluation metrics, A/B testing, security controls, and governance. Fills the Domain 5 gap completely.
Use these when you want a direct jump instead of following the full path.
These four guides are driven by your specific weak spots from 120 practice questions. Each one covers tips, gotchas, mental traps, and service decision rules — not wrong/right answer lists.
FM selection criteria, model distillation, BDA blueprints, embedding chunking strategies, vector stores, semantic cache, KB sync patterns.
Strands SDK + MCP, Bedrock Flows vs agents, ReAct vs CoT, SageMaker inference types, shadow testing, async pipelines, API Gateway patterns.
Guardrails component map, monitoring metrics/dimensions, IAM Identity Center vs IAM, VPC Interface endpoint, OIDC+Cognito, PII toolkit, BOLD dataset.
Batch inference for cost, semantic caching, CloudWatch GenAI observability, CreateEvaluationJob API, 1,000-prompt limit, Cognito workforce, troubleshooting decision table.