Advanced Generative AI Development on AWS Training & Certification Course

✦ Generative AI ✦ Amazon Bedrock ✦ Foundation Models ✦ Large Language Models ✦ LLM Application Development ✦ Prompt Engineering ✦ Advanced Prompt Engineering ✦ Retrieval-Augmented Generation ✦ RAG ✦ Vector Databases ✦ Embeddings ✦ Amazon Bedrock Knowledge Bases ✦ OpenSearch ✦ Agentic AI ✦ Amazon Bedrock AgentCore ✦ AI Agents ✦ Tool Integration ✦ Multimodal AI ✦ Data Processing ✦ AI Safety ✦ Responsible AI ✦ AI Security ✦ AI Governance ✦ Model Evaluation ✦ GenAI Testing ✦ GenAI Monitoring ✦ Observability ✦ Performance Optimization ✦ Cost Optimization ✦ Token Optimization ✦ Enterprise AI Architecture ✦ Cloud Architecture ✦ AWS AI Services ✦ API Integration ✦ Serverless Architecture ✦ Event-Driven Architecture ✦ Infrastructure as Code ✦ CI/CD for AI Applications

• ✓ Working knowledge of AWS Cloud services and architecture
• ✓ Experience developing or deploying applications
• ✓ Understanding of AI/ML concepts
• ✓ Familiarity with Generative AI and foundation models
• ✓ Basic understanding of Large Language Models (LLMs)
• ✓ Familiarity with APIs and application integration
• ✓ Understanding of cloud security fundamentals
• ✓ Basic knowledge of databases and data processing
• ✓ Familiarity with software development practices
• ✓ Experience with Python or another programming language is recommended
• ✓ Familiarity with containers, serverless technologies or cloud-native architectures is beneficial
• ✓ Prior hands-on experience with Amazon Bedrock is beneficial but not mandatory

Module 1: Generative AI Architecture & Production Readiness

• Generative AI landscape
• Foundation models and LLMs
• Generative AI application architectures
• Moving from experimentation to production
• Enterprise Generative AI requirements
• Scalability and reliability considerations
• Security and compliance considerations
• Selecting appropriate AWS services

Module 2: Foundation Model Selection & Integration

• Foundation model concepts
• Evaluating foundation models
• Model selection criteria
• Model benchmarking
• Performance evaluation
• Dynamic model selection
• Amazon Bedrock foundation models
• Foundation model integration patterns
• API-based model integration
• Multi-model architectures

Module 3: Advanced Data Processing & Multimodal AI

• Generative AI data requirements
• Data preparation
• Data validation
• Data processing pipelines
• Multimodal inputs
• Text, image and other data modalities
• Data quality considerations
• Data optimization
• Preparing enterprise data for GenAI applications

Module 4: Retrieval-Augmented Generation

• RAG architecture
• RAG workflow
• Document ingestion
• Chunking strategies
• Embeddings
• Semantic search
• Retrieval strategies
• Context augmentation
• Hybrid retrieval
• RAG evaluation
• RAG optimization

Module 5: Vector Databases & Knowledge Bases

• Vector database fundamentals
• Embeddings and similarity search
• Vector indexing
• Amazon Bedrock Knowledge Bases
• Amazon OpenSearch
• Hybrid retrieval architectures
• Metadata filtering
• Retrieval optimization
• Enterprise knowledge systems

Module 6: Advanced Prompt Engineering

• Prompt engineering fundamentals
• Advanced prompting strategies
• Prompt templates
• Prompt management
• Structured prompting
• Few-shot techniques
• Reasoning-oriented prompting
• Prompt evaluation
• Enterprise prompt governance
• Prompt security
• Managing prompt versions

Module 7: Agentic AI & Tool Integration

• Agentic AI concepts
• AI agent architecture
• Agent workflows
• Planning and reasoning
• Tool integration
• Function/API integration
• Multi-step workflows
• Agent orchestration
• Amazon Bedrock AgentCore
• Enterprise agent architectures
• Agent security and governance

Module 8: AI Safety, Security & Responsible AI

• Generative AI security
• Responsible AI
• Content filtering
• Privacy protection
• Data protection
• Prompt injection risks
• Adversarial testing
• Secure model integration
• Identity and access management
• Governance
• Compliance
• Enterprise AI security controls

Module 9: Performance & Cost Optimization

• GenAI performance optimization
• Token optimization
• Latency management
• Model selection for cost efficiency
• Intelligent caching
• Batching
• Throughput optimization
• Cost monitoring
• Cost-performance trade-offs
• Production optimization strategies

Module 10: Monitoring & Observability

• GenAI application monitoring
• Model monitoring
• Application performance monitoring
• Logging
• Metrics
• Tracing
• AI observability
• Quality monitoring
• Failure detection
• Production troubleshooting

Module 11: Testing & Validation

• GenAI testing strategies
• Model evaluation
• Application evaluation
• Response quality
• Accuracy and relevance
• Safety testing
• Regression testing
• Automated evaluation
• Continuous validation
• Troubleshooting GenAI applications

Module 12: Enterprise Integration & Production Deployment

• Enterprise GenAI architecture
• Application integration
• API integration
• Serverless architectures
• Event-driven architectures
• Secure application patterns
• Infrastructure as Code
• CI/CD for AI applications
• Production deployment
• Reliability
• Disaster recovery considerations
• Enterprise governance