AIP-C01 Study Guide: AWS Certified Generative AI Developer

How to Pass AWS Certified Generative AI Developer - Professional (AIP-C01) in 2026 — hero

The AIP-C01 exam (AWS Certified Generative AI Developer - Professional) costs USD 300, or JPY 40,000 in Japan. It has 75 questions in 180 minutes and passes at 750 out of 1,000. Foundation model integration, data management and compliance is the largest of its five domains at 31%.

Practice this exam in timed mode with the StudyKits AIP-C01 Generative AI Developer Professional app. The AWS certification hub lists every AWS exam guide and practice app.

It is a professional exam for people who already build applications on AWS and have hands-on generative AI experience; plan about 6 weeks. Exam facts are from the AWS pages under Sources, checked October 15, 2026.

Naming note: AIP-C01 is not the foundational AWS Certified AI Practitioner. That exam is AIF-C01 (90 minutes, USD 100) and is not covered here. This page’s URL contains “ai-practitioner” for link stability.

The beta ended on March 31, 2026. The standard version was refreshed to include Amazon Bedrock AgentCore. If you are new to AWS or to generative AI, start with AIF-C01 or with hands-on Amazon Bedrock work first.

What is the AIP-C01 exam format?

The AIP-C01 validates a candidate’s ability to effectively integrate foundation models (FMs) into applications and business workflows. Specifically, AWS expects you to be able to:

Design and implement solutions using vector stores, Retrieval-Augmented Generation (RAG), knowledge bases, and other GenAI architectures
Integrate foundation models into applications and business workflows
Apply prompt engineering and prompt management techniques
Implement agentic AI solutions
Optimize GenAI applications for cost, performance, and business value
Implement security, governance, and Responsible AI practices
Troubleshoot, monitor, and optimize GenAI applications in production
Evaluate foundation models for quality and responsibility

Exam format:

Questions: 75 total — 65 scored, 10 unscored (unscored questions are not identified during the exam)
Time: 180 minutes (3 hours)
Passing score: 750 out of 1,000 (scaled)
Cost: USD 300 (JPY 40,000)
Delivery: Pearson VUE testing centers or online proctoring

Recommended experience: 2+ years building production-grade applications on AWS or with open-source technologies, general AI/ML or data engineering experience, and at least 1 year of hands-on experience implementing generative AI solutions. AWS also expects working knowledge of AWS compute, storage, networking, security, deployment/IaC tooling, monitoring, and cost optimization.

Model development, model training, advanced ML techniques, and data/feature engineering are explicitly out of scope — this exam tests integration and application-building skill, not building models from scratch.

What are the five AIP-C01 domains?

Content Domain 1: Foundation Model Integration, Data Management, and Compliance (31%)

The largest domain (31%, five points ahead of Domain 2). Covers selecting and integrating foundation models from Amazon Bedrock (Anthropic, Meta, Amazon Titan, Mistral, and others), designing RAG architectures with vector stores and knowledge bases, data pipeline design for GenAI applications, and compliance considerations for the data those applications touch.

Content Domain 2: Implementation and Integration (26%)

Covers the practical mechanics of wiring foundation models into applications: API integration patterns with Bedrock, building agentic workflows, orchestrating multi-step GenAI pipelines, and integrating with other AWS services (Lambda, Step Functions, API Gateway) to deliver a working application, not just a model call.

Content Domain 3: AI Safety, Security, and Governance (20%)

Covers Bedrock Guardrails, content filtering and PII redaction, IAM and data security for GenAI workloads, Responsible AI evaluation, and governance processes — model versioning, approval workflows, and auditability for production GenAI systems.

Content Domain 4: Operational Efficiency and Optimization for GenAI Applications (12%)

Covers cost optimization specific to GenAI workloads (token-based pricing, provisioned throughput vs on-demand, model selection tradeoffs), performance tuning, and scaling strategies for production inference.

Content Domain 5: Testing, Validation, and Troubleshooting (11%)

Covers evaluating foundation model outputs for quality and correctness, testing strategies for non-deterministic GenAI systems, monitoring GenAI applications in production, and troubleshooting common failure modes (hallucination, latency, cost spikes).

Which AWS services should you know?

Amazon Bedrock — the core service. Know model access, Agents, Knowledge Bases, Guardrails, provisioned throughput vs on-demand, and model evaluation.
Amazon SageMaker — JumpStart for pre-trained models, Studio, and how SageMaker complements Bedrock for custom model work (even though model training itself is out of scope for this exam).
Amazon Q Business and Amazon Q Developer — enterprise AI assistant and AI coding assistant, including how they fit into a GenAI application portfolio.
Vector and retrieval infrastructure — Bedrock Knowledge Bases, OpenSearch Serverless as a vector store, and RAG pipeline design end to end.
Orchestration and integration — Lambda, Step Functions, API Gateway, and EventBridge as the glue around foundation model calls.
Security and governance tooling — Bedrock Guardrails, IAM for Bedrock resource access, CloudTrail for auditing model invocations, and Macie for sensitive data discovery in GenAI data pipelines.

What does a 6-week study plan look like?

This plan assumes you already have hands-on AWS development experience and are studying the GenAI-specific content, roughly 1.5-2 hours per day, five days a week.

Week 1: Foundation Model Integration Fundamentals

Study Bedrock model access, model families and selection criteria, and RAG architecture basics (embeddings, vector stores, retrieval, generation). Hands-on: stand up a minimal RAG pipeline using Bedrock Knowledge Bases against a small document set. 2 practice question sets.

Week 2: Data Management and Compliance for GenAI

Study data pipeline design for GenAI applications, data governance and compliance considerations, and knowledge base ingestion strategies. Hands-on: configure a Knowledge Base data source with a defined chunking strategy. 2 practice question sets.

Week 3: Implementation and Integration

Study agentic workflow design with Bedrock Agents, orchestration with Step Functions and Lambda, and multi-step GenAI pipeline patterns. Hands-on: build a Bedrock Agent that calls an external action group. 3 practice question sets.

Week 4: AI Safety, Security, and Governance

Study Bedrock Guardrails configuration, IAM policies scoped to Bedrock resources, PII redaction, and model governance/versioning workflows. Hands-on: apply a Guardrail with content filters and topic restrictions to an existing Bedrock application. 2 practice question sets.

Week 5: Operational Efficiency, Testing, and Troubleshooting

Study cost optimization (token pricing, provisioned throughput decisions), performance tuning, model output evaluation techniques, and troubleshooting production GenAI issues. Hands-on: compare on-demand vs provisioned throughput cost for a sample workload. 2 practice question sets.

Week 6: Review and Exam Readiness

Take a full-length practice exam (75 questions, timed to 180 minutes). Identify your weakest domains and re-study them. Take a second full-length practice exam. Aim for 80%+ before sitting the real exam.

How should you practise for AIP-C01?

AWS Skill Builder has official AIP-C01 practice question sets, also in Japanese, Korean and Simplified Chinese. The StudyKits AIP-C01 app has original question sets covering all five domains, with explanations for every answer choice.

Start early. Begin practice questions from the end of Week 1, not just during final review — early exposure teaches you what AWS actually tests versus what you assume they test.

Review every explanation, including questions you answered correctly. Getting the right answer for the wrong reason is a gap that will catch up with you on a harder question later.

Track your scores by domain. If you consistently score below 70% on the AI Safety/Governance domain, that is where you need to focus next.

Simulate exam conditions at least twice during preparation — a full question set, timed, no notes. 180 minutes for 75 questions is more forgiving than it sounds once questions get long and scenario-based, but stamina still matters.

What should you do on exam day?

The night before: Stop studying by 8 PM. Review your notes one final time, then close the books. Get at least 7 hours of sleep.

During the exam: Read every question completely before looking at the answers. Watch for qualifiers like “most cost-effective” or “with the least operational overhead” — they change which answer is correct. Flag uncertain questions and return to them; 180 minutes for 75 questions gives you roughly 2.4 minutes per question, similar pacing to other AWS professional-level exams.

Elimination strategy: Eliminate answers that suggest anti-patterns first — unnecessary complexity, ignoring a stated constraint, or using a service outside its intended purpose.

What comes after AIP-C01?

AIP-C01 sits at the professional level, so the natural next moves are lateral rather than upward:

AWS Machine Learning Engineer - Associate (MLA-C02 general availability announced for January 14, 2027; the exam page lists TBD) if you want to move from integrating foundation models to building and training custom ML models
AWS Solutions Architect Professional (SAP-C03 replaces SAP-C02 on November 17, 2026) to pair GenAI implementation skills with broader architecture authority
Deeper Bedrock specialization through continued hands-on work with Agents, Guardrails, and multi-model orchestration

AIP-C01 tests whether you can ship production generative AI applications on AWS, not just describe foundation models. Follow the 6-week plan and practise full 75-question sets under time.

Frequently asked questions

How much does the AWS generative AI certification cost?

The AIP-C01 exam costs USD 300 (JPY 40,000). The foundational AI Practitioner (AIF-C01) costs USD 100.

What is the AIP-C01 passing score?

750 on a scale of 100-1,000.

Is AIP-C01 still in beta?

No. The beta ended on March 31, 2026, and the standard exam is open for registration.

What is the difference between AIF-C01 and AIP-C01?

AIF-C01 (AI Practitioner) is foundational and does not expect you to build. AIP-C01 (Generative AI Developer - Professional) expects you to build and run generative AI applications on AWS.

Which languages is AIP-C01 offered in?

English, Japanese, Korean and Simplified Chinese.

Keep reading

AIP-C01 4-Week Study Plan for Working Developers

Sources

Checked October 15, 2026:

How to Pass AWS Certified Generative AI Developer - Professional (AIP-C01) in 2026