Generative AI Developer Professional mindmap

AWS Certified Generative AI Developer Professional (AIP-C01)

Validates integrating foundation models into applications and business workflows in production on AWS

• Domain 1: Foundation Model Integration, Data Management, and Compliance (31%): Design solutions, pick models, prepare data, build vector stores and retrieval, manage prompts
  • 1.1 Analyze requirements and design GenAI solutions: Architectures that match business needs and technical constraints
  • 1.2 Select and configure FMs: Choose models by benchmarks, capabilities and limitations; switch models without code changes
  • 1.3 Implement data validation and processing pipelines for FM consumption: Quality checks and processing for text, image, audio and tabular data
  • 1.4 Design and implement vector store solutions: Vector databases for semantic retrieval beyond traditional search
  • 1.5 Design retrieval mechanisms for FM augmentation: Document segmentation, embeddings and retrieval configuration
  • 1.6 Implement prompt engineering strategies and governance for FM interactions: Instruction frameworks, prompt management and responsible AI guidelines
• Domain 2: Implementation and Integration (26%): Agents, deployment, enterprise integration, FM APIs and development tools
  • 2.1 Implement agentic AI solutions and tool integrations: Agents with memory and state, multi-agent systems and MCP tool access
  • 2.2 Implement model deployment strategies: On-demand, provisioned and hybrid deployment of FMs
  • 2.3 Design and implement enterprise integration architectures: Connect FM capabilities to existing enterprise systems
  • 2.4 Implement FM API integrations: Synchronous, asynchronous and real-time model interaction
  • 2.5 Implement application integration patterns and development tools: API interfaces for GenAI workloads, and tools that speed adoption
• Domain 3: AI Safety, Security, and Governance (20%): Input and output controls, data security, governance and responsible AI
  • 3.1 Implement input and output safety controls: Block harmful inputs and prevent harmful outputs
  • 3.2 Implement data security and privacy controls: Protected environments and privacy-preserving systems
  • 3.3 Implement AI governance and compliance mechanisms: Compliance frameworks, decision logs and data source tracking
  • 3.4 Implement responsible AI principles: Transparent outputs and fairness evaluations
• Domain 4: Operational Efficiency and Optimization for GenAI Applications (12%): Cost, performance and monitoring
  • 4.1 Implement cost optimization and resource efficiency strategies: Token efficiency and cost-effective model selection
  • 4.2 Optimize application performance: Latency and cost trade-offs, streaming and retrieval performance
  • 4.3 Implement monitoring systems for GenAI applications: Observability for FM interactions and GenAI-specific indicators
• Domain 5: Testing, Validation, and Troubleshooting (11%): Evaluation systems and troubleshooting GenAI applications
  • 5.1 Implement evaluation systems for GenAI: Assessment frameworks and systematic model evaluation
  • 5.2 Troubleshoot GenAI applications: Content handling issues and FM integration issues