All 5 domains
20 official task statements
95–133 hrs suggested study
-
Domain 01
31%
The heaviest AIP-C01 domain covers designing GenAI solutions, choosing and customizing foundation models, preparing data for them, building vector stores and retrieval, and governing prompts, mostly with Amazon Bedrock, SageMaker AI and OpenSearch Service.
Inside this domain
- Analyze requirements and design GenAI solutions
- Select and configure FMs
- Implement data validation and processing pipelines for FM consumption
- Design and implement vector store solutions
- and 2 more task statements
Watch for Fine-tune for fresh facts
- Task statements
- 6
Read the domain guide →
-
Domain 02
26%
This domain covers building agentic systems with tools and MCP, deploying foundation models, connecting GenAI to enterprise systems and CI/CD, integrating Bedrock APIs reliably, and using AWS developer tools to build and troubleshoot GenAI applications.
Inside this domain
- Implement agentic AI solutions and tool integrations
- Implement model deployment strategies
- Design and implement enterprise integration architectures
- Implement FM API integrations
- and 1 more task statements
Watch for Agent will stop itself
- Task statements
- 5
Read the domain guide →
-
Domain 03
20%
This domain covers protecting GenAI applications end to end: filtering harmful inputs and outputs, reducing hallucinations, detecting prompt injection, protecting sensitive data, tracking lineage for compliance and applying responsible AI principles such as transparency and fairness.
Inside this domain
- Implement input and output safety controls
- Implement data security and privacy controls
- Implement AI governance and compliance mechanisms
- Implement responsible AI principles
Watch for System prompt as security
- Task statements
- 4
Read the domain guide →
-
Domain 04
12%
This domain covers running GenAI applications efficiently: reducing token spend, choosing cost-effective models, caching, cutting latency, raising throughput, tuning inference parameters and monitoring token usage, quality, tools and vector stores in production.
Inside this domain
- Implement cost optimization and resource efficiency strategies
- Optimize application performance
- Implement monitoring systems for GenAI applications
Watch for Bigger model for everything
- Task statements
- 3
Read the domain guide →
-
Domain 05
11%
The smallest AIP-C01 domain covers evaluating GenAI output quality, comparing models, testing retrieval and agents, validating deployments, and troubleshooting context window, API, prompt and retrieval problems in foundation model applications.
Inside this domain
- Implement evaluation systems for GenAI
- Troubleshoot GenAI applications
Watch for Accuracy for free text
- Task statements
- 2
Read the domain guide →
Domain names, weights and task statements come from Amazon Web Services's official outline (AIP-C01 exam guide (current, generally available)). Question counts and study hours are editorial planning estimates, not official figures. Last reviewed October 2026.