Answer in brief
AWS Certified Generative AI Developer – Professional (AIP-C01) is a professional-level exam on building production generative AI applications on AWS. It has 75 multiple-choice and multiple-response questions (65 scored, 10 unscored) in 180 minutes. It costs $300 and passes at a scaled 750 out of 1,000.
- What it tests: integrating foundation models (FMs) into applications and business workflows. That means RAG, vector stores, agents, prompt engineering, safety, cost, monitoring and evaluation.
- Domains: five. The largest is Foundation Model Integration, Data Management, and Compliance at 31%. Next are Implementation and Integration (26%) and AI Safety, Security, and Governance (20%).
- Who it is for: developers with production experience. AWS recommends 2 or more years building production-grade applications and 1 year of hands-on generative AI work.
- What it does not test: model training, advanced ML techniques, and data and feature engineering.
- How to prepare: read the official exam guide, take a diagnostic by domain, fix the heaviest weak domains first, build one real Amazon Bedrock application, and book when practice results are strong in every domain.
Everything in this guide comes from the official AWS exam guide and the AWS certification page unless I say otherwise.
See the study system in action
How StudyTech AI works
Watch how your answers become a focused daily plan, targeted practice and flashcards. See how StudyTech AI tracks your progress before you decide when to book.
Ready to explore? Visit StudyTech AI to see supported certifications and choose your next step.
Explore StudyTech AIThe exam at a glance
75 questions, 180 minutes, $300, 750 to pass, no prerequisites.
| Detail | AWS Certified Generative AI Developer – Professional |
|---|---|
| Exam code | AIP-C01 |
| Level | Professional |
| Questions | 75 (65 scored, 10 unscored) |
| Question types | Multiple choice (one correct of four) and multiple response (two or more correct of five or more) |
| Time | 180 minutes |
| Price | $300 USD |
| Passing score | 750 on a scaled range of 100–1,000 |
| Scoring model | Compensatory: you pass the exam overall, not each domain |
| Delivery | Pearson VUE test centre or online proctored |
| Languages | English, Japanese, Korean and Simplified Chinese |
| Prerequisites | None required |
| Validity | 3 years; recertify by passing the latest version of the exam |
Three details worth knowing before you book:
- Unscored questions are hidden. Ten of the 75 do not count, but you cannot tell which. Treat every question as scored.
- Blank answers are wrong. There is no penalty for guessing, so never leave a question empty.
- Section feedback is a guide, not a verdict. Your score report may show how you did per section. AWS itself says to use caution when you read it.
The five domains and their weights
Domain 1 is almost a third of the scored content; Domains 1 and 2 together are 57%.
| # | Domain | Weight | Rough scored questions (of 65) |
|---|---|---|---|
| 1 | Foundation Model Integration, Data Management, and Compliance | 31% | ~20 |
| 2 | Implementation and Integration | 26% | ~17 |
| 3 | AI Safety, Security, and Governance | 20% | ~13 |
| 4 | Operational Efficiency and Optimization for GenAI Applications | 12% | ~8 |
| 5 | Testing, Validation, and Troubleshooting | 11% | ~7 |
The weights are AWS's. The "rough scored questions" column is my own arithmetic (weight × 65), not an AWS figure. Use it to plan effort, not as a count.
The services named below are the examples AWS gives in each domain's skill list. AWS says the guide is not a complete list of exam content, so treat them as the core, not the edge.
Domain 1: Foundation Model Integration, Data Management, and Compliance (31%)
What it tests: designing a generative AI solution and getting the right data to the right model. It has six tasks:
- Analyse requirements and design GenAI solutions
- Select and configure FMs
- Implement data validation and processing pipelines for FM consumption
- Design and implement vector store solutions
- Design retrieval mechanisms for FM augmentation
- Implement prompt engineering strategies and governance for FM interactions
Services that matter: Amazon Bedrock (including Cross-Region Inference, Knowledge Bases, Guardrails, Prompt Management, Prompt Flows and reranker models), Amazon Titan embeddings, Amazon OpenSearch Service, Amazon Aurora with pgvector, Amazon SageMaker AI and SageMaker Model Registry, AWS Lambda, AWS Step Functions, AWS Glue Data Quality, Amazon Comprehend, Amazon DynamoDB and Amazon S3. The AWS Well-Architected Generative AI Lens also appears.
This is the RAG domain. Expect questions on chunking, embedding choice, hybrid search, keeping vector stores current, and model choice and resilience across Regions. MCP clients for vector queries are named here too.
Domain 2: Implementation and Integration (26%)
What it tests: building and wiring up the application. Five tasks:
- Implement agentic AI solutions and tool integrations
- Implement model deployment strategies
- Design and implement enterprise integration architectures
- Implement FM API integrations
- Implement application integration patterns and development tools
Services that matter: Amazon Bedrock APIs (including streaming, provisioned throughput, Prompt Flows and Bedrock Data Automation), Strands Agents and AWS Agent Squad, Model Context Protocol (MCP) servers on Lambda or Amazon ECS, Step Functions, Amazon API Gateway, Amazon EventBridge, Amazon SQS, SageMaker AI endpoints, AWS CodePipeline and AWS CodeBuild, AWS X-Ray, AWS Amplify and Amazon Q Developer.
This is the agents domain. Expect stopping conditions, timeouts and circuit breakers for agent workflows, human review steps, retries with exponential backoff, and model routing.
Domain 3: AI Safety, Security, and Governance (20%)
What it tests: keeping inputs, outputs and data safe, and proving it. Four tasks:
- Implement input and output safety controls
- Implement data security and privacy controls
- Implement AI governance and compliance mechanisms
- Implement responsible AI principles
Services that matter: Amazon Bedrock Guardrails, Amazon Bedrock Knowledge Bases (for grounding responses), Amazon Comprehend and Amazon Macie for PII detection, VPC endpoints, IAM policies, AWS Lake Formation, SageMaker AI model cards, AWS Glue Data Catalog, AWS CloudTrail and Amazon CloudWatch Logs.
Expect prompt injection and jailbreak defence, defence in depth (pre-processing, guardrails, post-processing), PII masking, audit trails and data lineage.
Domain 4: Operational Efficiency and Optimization for GenAI Applications (12%)
What it tests: cost, speed and monitoring. Three tasks:
- Implement cost optimisation and resource efficiency strategies
- Optimise application performance
- Implement monitoring systems for GenAI applications
Services that matter: Amazon CloudWatch, Amazon Bedrock Model Invocation Logs, latency-optimised Amazon Bedrock models and provisioned throughput. Much of this domain is about techniques rather than services: token tracking, prompt and semantic caching, tiered model use by query complexity, batching, streaming, and temperature and top-k/top-p settings.
Domain 5: Testing, Validation, and Troubleshooting (11%)
What it tests: proving the application works, and fixing it when it does not. Two tasks:
- Implement evaluation systems for GenAI
- Troubleshoot GenAI applications
Services that matter: Amazon Bedrock Model Evaluations and Amazon Bedrock Agent evaluations, CloudWatch Logs and AWS X-Ray. Expect LLM-as-a-judge, RAG and retrieval quality testing, A/B and canary testing, regression tests on model output, and diagnosing context window overflow, chunking problems and embedding quality.
What is out of scope
AWS lists three job tasks the target candidate is not expected to do:
- Model development and training
- Advanced ML techniques
- Data engineering and feature engineering
If a resource spends most of its time on training models from scratch, it is preparing you for a different exam.
Who should take it
Developers who already ship production applications and have built with generative AI for real.
AWS describes the target candidate as having:
- 2 or more years of experience building production-grade applications on AWS or with open-source technologies
- General AI/ML or data engineering experience
- 1 year of hands-on experience implementing GenAI solutions
AWS also recommends this AWS knowledge:
- AWS compute, storage and networking services
- AWS security best practices and identity management
- AWS deployment and infrastructure as code (IaC) tools
- AWS monitoring and observability services
- AWS cost optimisation principles
There are no required prerequisites. The certification page notes that candidates may benefit from holding AWS Certified AI Practitioner, Solutions Architect – Associate, Machine Learning Engineer – Associate or Data Engineer – Associate first. That is a suggestion, not a requirement.
If you are newer to AWS, an Associate exam first is the sensible order. I cover one option in is the AWS Machine Learning Engineer – Associate worth it.
How to prepare: a step-by-step plan
Start with a diagnostic by domain, fix the heaviest weak areas first, build one real Bedrock app, then book on evidence.
- Read the official exam guide end to end. Every domain page lists tasks and skills. Use it as a checklist and mark each skill as strong, shaky or unknown.
- Take a diagnostic before any course. Find out which of the five domains are weak. Time spent on what you already know is time taken from a gap.
- Prioritise by weight × weakness. A weak Domain 1 (31%) costs you far more than a weak Domain 5 (11%). But scoring is overall, so do not leave any domain badly weak.
- Build one small application that touches every domain. For example: a Bedrock Knowledge Base over your own documents, an agent with one tool, Guardrails on input and output, invocation logging to CloudWatch, and a simple evaluation run. Scenario questions are much easier once you have hit the problems for real.
- Learn the trade-offs, not just the services. Professional questions stack constraints: cost, latency, data residency and security in one stem. Know why you would pick provisioned throughput over on-demand, or a reranker over a bigger chunk size.
- Practise with scenario questions and review every wrong answer. Write down the domain and the constraint you missed. Turn each mistake into a flashcard and revisit it until it sticks.
- Book when every domain is consistently strong in practice. My rule of thumb is around 80% per domain, held over several sessions. That is a personal heuristic, not an AWS figure and not a guarantee.
Prepare with StudyTech AI
StudyTech AI is a personalised AWS and AI certification prep platform. It finds your knowledge gaps, builds your practice around your skills and gaps, turns your mistakes into flashcards, scores your readiness for every exam domain, and tells you when you are ready to take your exam.
For AIP-C01, that means you start with a diagnostic across all five domains and spend your time on the weakest ones first.
Pricing is $29 a month or $297 a year, and the diagnostic is free.
- Try 15 free AIP-C01 practice questions, mapped to the five domains with full explanations
- See the AIP-C01 certification page for the full exam overview
Start a free AIP-C01 diagnostic on StudyTech AI
Cite this page
StudyTech AI. "What Is on the AWS Generative AI Developer – Professional (AIP-C01) Exam? Domains, Format and How to Prepare (2026)." studytech.ai/blog/aws-generative-ai-developer-professional-exam-guide. Updated 8 November 2026.
Related reading
- AI certifications on StudyTech AI
- The best AWS practice exams in 2026
- Is the AWS Machine Learning Engineer – Associate worth it in 2026?
- The best AI certifications: a 2027 tier list
Verified 8 November 2026 against the official AWS Certified Generative AI Developer – Professional (AIP-C01) exam guide, its five domain pages, the AWS certification page and the AWS recertification page. AWS can update the exam guide, so confirm the current version before you book.