Implement model evaluation frameworks

Design flexible routing strategies

Build resilient AI systems with AWS services

Build effective data validation and processing pipelines for foundation models

Identify required and optional JSON fields for different foundation models

Design synchronized audio-visual workflows using Amazon Transcribe

Design and implement vector store solutions

Create workflows that automatically sync source data with vector stores

Evaluate governance strategies to implement effective version control and deployment practices

Design and implement effective retrieval mechanisms

Manage Knowledge Base updates through incremental ingestion and synchronization

Construct end-to-end RAG systems by applying standardized interfaces that enforce security controls

Implement enterprise foundation model control systems with Amazon Bedrock

Implement validation systems for probabilistic foundation model outputs

Apply strategies that integrate advanced prompting with business processes while maintaining governance

Build a question-answering application using Amazon Bedrock Knowledge Bases

Use the Retrieve and RetrieveAndGenerate APIs to query a knowledge base

Run through notebooks that answer questions from enterprise knowledge bases

Implement Strands Agent modular patterns for agent development

Describe the core benefits of model extension frameworks for foundation model enhancement

Implement a text summarization workflow using Strands Agents, AgentCore, and MCP

Describe the benefits and use cases for model deployment strategies

Design model selection frameworks for generative AI applications

Develop safeguarding mechanisms to protect your AI workflows

Describe the core principles of enterprise integration architectures for AI systems

Describe the benefits of event-driven processing in generative AI applications

Implement cross-environment generative AI solutions

Describe the core principles of foundation model API integrations for AI systems

Explain the benefits of various streaming patterns for conversational AI

Implement conversational AI systems using WebSockets and Server-sent events

Enhance enterprise systems by using AI-assisted AWS services

Create automated content enhancement and transformation workflows

Use spec-driven development features for complex projects

Use the Amazon Nova Lite model through Amazon Bedrock API for intelligent question answering

Identify the limitations of zero-shot prompting and enhance response accuracy by providing context

Implement both complete and streaming response generation methods for RAG-based systems

Describe purpose and importance of content input safety controls

Identify common techniques for input filtering

List best practices for implementing content input safety

Describe defense-in-depth for AI strategy and the seven layer approach

Locate resources for AI security best practices

Implement secure infrastructure for AI data processing with network isolation using VPC endpoints

Define concepts of compliance, governance, and security

Locate AWS resources for compliance programs

Use AWS services and tools to implement compliance and governance controls for generative AI applications

Enable model access in Amazon Bedrock

Set up a knowledge base using provided sample files in an S3 bucket

Test the knowledge base with sample prompts

Implement token estimation and tracking mechanisms to optimize foundation models

Design capacity planning strategies for foundation model deployments on AWS

Apply monitoring and maintenance best practices for intelligent caching systems

Implement pre-computation strategies to reduce query response times

Implement Amazon Bedrock AgentCore services to accelerate agent deployment

Monitor and analyze API performance metrics to identify optimization opportunities

Implement comprehensive monitoring and observability systems for generative AI applications

Implement call pattern tracking systems that identify usage trends and anomalies

Apply optimization techniques for FAISS indexes and approximate nearest neighbor algorithms

Assess generative AI outputs using metrics for relevance, factual accuracy, consistency, and fluency

Design quality assurance processes integrating continuous evaluation workflows and regression testing

Apply testing methods including A/B testing and multi-model comparison for output quality

Describe common issues and challenges of generative AI

Describe methods for resolving content handling issues

Describe methods for resolving retrieval system issues