AIP-C01 AWS Generative AI Developer – Professional
- Vendor
- AWS
- Level
- Professional
- Questions
- 75
- Time
- 180 min
- Domains
- 5
- In the bank
- 465
Practice 465 questionsTake a timed mock examDay-by-day study plan
Guides for this exam
- Prompting, RAG, Fine-Tuning or Continued Pre-Training: One Rule That Works on AWS, Azure, Google Cloud and Databricks Exams · 한국어
- AIP-C01 Study Guide: Four Diagnostic Questions That Sort Almost Every Generative AI Developer Scenario · 한국어
The exam at a glance
| Level | Professional |
|---|---|
| Questions | 75 |
| Time | 180 minutes (about 2.4 minutes per question) |
| Passing mark | 750 on a scaled score of 100–1,000 |
Domains and weights
| Domain | Weight | Practice questions here |
|---|---|---|
| D1 · Foundation Model Integration, Data Management, and Compliance | 31% | 134 |
| D2 · Implementation and Integration | 26% | 113 |
| D3 · AI Safety, Security, and Governance | 20% | 117 |
| D4 · Operational Efficiency and Optimization for GenAI Applications | 12% | 53 |
| D5 · Testing, Validation, and Troubleshooting | 11% | 48 |
59 of the 465 practice questions (13%) ask for more than one answer.
D1 · Foundation Model Integration, Data Management, and Compliance 31%
Much of this domain concerns the data side of a GenAI system: chunking, embeddings, vector stores, knowledge bases and model selection. Scenarios describe a quality, cost or access requirement and ask which design meets it. Good answers enforce rules in the data layer and keep the index in step with its sources, while weak ones lean on prompt instructions or application code.
What to watch for
- First decide whether the need is new knowledge or new behavior. Missing facts call for RAG; a consistent tone or format calls for fine-tuning.
- An index drifts unless edits and deletions in the source reach it. A new embedding size means the whole index must be rebuilt, or stored vectors and new queries will no longer line up.
- For multilingual requirements, test candidate models with real questions in every target language before committing to one.
Tasks in the official exam guide (6)
D2 · Implementation and Integration 26%
Implementation scenarios put a GenAI feature into production: agents calling tools, partners calling APIs, models that throttle or slow down, and new code that must ship safely. You choose the deployment, integration or resilience pattern. The best answers limit damage and cost together, so watch for options that retry forever, pay for idle capacity or roll changes out all at once.
What to watch for
- When a dependency keeps failing, prefer bounded retries with backoff and a way to stop calling it, rather than longer or endless retries.
- Match the deployment to the traffic: steady, predictable load and intermittent load favor different options, and idle cost often decides between them.
- If a smaller specialized model meets both the accuracy and latency targets, serving it usually beats tuning the price of a much larger one.
Tasks in the official exam guide (5)
D3 · AI Safety, Security, and Governance 20%
Safety and governance questions mix runtime controls with documentation and policy. A scenario may involve sensitive data in responses, a topic that must be blocked everywhere, prompt injection or a regulated use case, and you pick the mechanism that fits. Classify the problem first (safety, privacy, governance or responsible AI), because distractors often come from the wrong family.
What to watch for
- Guardrail settings are specific: sensitive information is blocked or masked per entity type, and topic, grounding and format checks each solve a different problem.
- Separate enforcement from documentation. Guardrails act on each request, while SageMaker AI Model Cards record a model's purpose, evaluations and limits.
- For streamed responses, weigh screening before display against speed; the processing modes trade latency for the guarantee that nothing unchecked reaches the screen.
Tasks in the official exam guide (4)
D4 · Operational Efficiency and Optimization for GenAI Applications 12%
Optimization questions start from a symptom: slow ingestion, a rising error rate, a cost that must drop, or a finance request for usage per application. The task is to find the cause or the simplest mechanism that fixes it. Decide first whether the problem is about cost, performance or missing visibility, then look for the option aimed at that specific issue.
What to watch for
- Use distinct metrics to separate causes. Throttling and malformed client requests show up in different CloudWatch metrics and need different fixes.
- Route requests by difficulty instead of choosing one model for everything; simple work can go to a cheaper model while hard cases keep the stronger one.
- For agents, step-level traces show where time and failures occur, which summary logs and bills cannot. Usage per application needs its own tagged inference profile.
Tasks in the official exam guide (3)
D5 · Testing, Validation, and Troubleshooting 11%
Testing questions ask how to prove a GenAI system works and how to find what broke when it does not. Expect scenarios about scoring open-ended answers, evaluating RAG, catching invented facts and explaining a drop after a change. Correct options measure the specific part in question; wrong ones rely on aggregate numbers, text overlap or guesses about which stage is slow.
What to watch for
- Evaluate RAG in two halves: whether the retriever found relevant context, and whether the generated answer stays faithful to it.
- Measure before tuning. Tracing per stage, or metrics per tool, shows where a slowdown or failure actually starts.
- For open-ended answers with no single reference, a rubric-based model judge works at scale, as long as humans check its scores from time to time.
A study order that works
The 35-day study plan for AIP-C01 takes the domains in this order. Each day has a short reading, key terms and a quiz from the bank.
- Days 1–10: D1 · Foundation Model Integration, Data Management, and Compliance
- Days 11–19: D2 · Implementation and Integration
- Days 20–26: D3 · AI Safety, Security, and Governance
- Days 27–30: D4 · Operational Efficiency and Optimization for GenAI Applications
- Days 31–34: D5 · Testing, Validation, and Troubleshooting
- Day 35: Final Mock Exam and Cross-Domain Review
Recent changes to these questions
- : 2 answer choice sets changed, 2 answer keys corrected, 2 explanations rewritten
- : 60 questions added
- : 4 explanations rewritten
- : 2 answer choice sets changed, 2 answer keys corrected, 2 explanations rewritten
- :
About these questions
Every question is original, written from the public exam guide and checked against the vendor's own documentation. None comes from a real exam. Each one has an explanation of why the answer is right and why the other options are wrong. How the questions are written.