Architecting and Implementing Generative AI Solutions in the Cloud
Contact us to book this course
Curriculum
Generative AI
Delivery methods
On-Site, Virtual
Duration
3 days
This advanced course is tailored for professionals keen on mastering generative AI (GenAI) within enterprise and cloud contexts. It begins with an introduction to the fundamentals of GenAI, including large language and multimodal models, setting the stage for their application in business. This course covers critical platforms and tools such as OpenAI GPT, Google Gemini, and the major cloud providers (AWS, Azure, and Google Cloud), with an emphasis on building and deploying generative AI applications that go beyond simple proofs of concept.
The curriculum progresses into specialized applications such as enterprise search, document summarization, Q&A systems, and constructing sophisticated enterprise chatbots, with a strong focus on responsible AI principles. By the end, participants will acquire the expertise to develop and implement cutting-edge GenAI solutions responsibly and effectively.
Learning objectives
- Build and deploy secure, safe, enterprise-grade GenAI applications
- Move beyond simple content generation into Enterprise Search, Large Document Summarization, Q&A, and chatbots
- Make informed choices regarding foundational models and cloud providers
- Improve large language models (LLM) results with advanced prompting engineering techniques and model fine-tuning
- Program GenAI solutions with Python and LangChain
- Leverage embeddings to improve search performance
- Implement the MapReduce, Refine, and Stuffing patterns for document summarization
- Ground results and avoid hallucinations by leveraging Retrieval Augmented Generation (RAG) solutions using cloud-based and open-source tools and services
- Deploy chatbots that work
- Ensure your AI solutions are safe using responsible AI principles and practices
Who should attend
- ML Practitioners
- Solution Developers
- Programmers integrating AI into their applications
- Cloud Architects and Engineers
Prerequisites
- Basic knowledge of GenAI and machine learning
- Programming experience with Python or another modern programming language
- Experience architecting and deploying systems on a major cloud provider (AWS, Azure, Google Cloud)
Course outline
- Architecting Reliable Solutions
- Understanding Regions and Zones
- Fault Tolerance
- Disaster Recovery
- Load Balancing
- Health Monitoring
- Activity: Architecting for Fault Tolerance
- Disaster Planning
- Analyzing Disaster Scenarios
- Risk Analysis
- Disaster Recovery Plans
- Activity: Disaster Planning
- Architecting for Scalability and Performance
- Vertical vs. Horizontal Scaling
- Distributed Computing
- Elastic Scalability
- Dealing with Latency
- Performance
- Activity: Architecting for Scalability and Performance
- Linux Basics
- Distributions
- SSH
- Bash Shell
- File System Commands
- Installing Apps with Apt or Yum
- Running Startup Scripts
- Infrastructure as a Service
- Extending Data Centers into the Cloud
- Advantages of IaaS
- VCPs
- VMs
- Exercise: Deploying Apps to VMs
- Platform as a Service
- PaaS Solutions
- Advantages of PaaS
- Leverage PaaS for Automated Deployments
- Serverless Services Options
- Pros and Cons of Serverless Services
- Exercise: Deploying Apps to PaaS
- Understanding Cost
- Case Studies
- Activity: Choosing Compute Services
- Portability
- Hybrid Cloud
- Multi-Cloud
- Cloud-Native Development
- Containers
- Advantages of Containers
- Docker
- Building Docker Images
- Running Containers
- Exercise: Building Docker Images
- Container Orchestration
- Kubernetes
- Advantages of Kubernetes
- Clusters
- Kubernetes Configuration
- Exercise: Kubernetes
- Storage Considerations
- Traditional vs. Cloud Data Considerations
- Storage Types Overview
- Availability
- Durability
- Structured vs. Unstructured Data
- Data Storage
- Binary Storage
- Block Storage
- NAS Services
- Data Warehousing
- Data Archiving
- Databases
- Overview of Databases
- Relational Databases
- NoSQL Databases
- Horizontal vs. Vertical Scaling
- Distributed Databases
- Strong vs. Eventual Consistency
- Failover Replicas
- Read Replicas
- Managed Relational and NoSQL Databases
- Understanding Storage Costs
- Activity: Evaluating Data Storage Solutions
- Securing Cloud Data
- Networking Considerations
- Controlling Access
- Scalability
- Stateless Design
- Load Balancing
- Elastic Scalability
- Exercise: Autoscaling and Load Balancing
- Designing for Reliability
- Fault-Tolerance
- Dealing with Failure
- Health Checkers and Autohealing
- Message Queueing
- Version Management
- Deploying New Versions
- Rolling Updates
- Blue/Green Deployments
- Canary Releases
- Exercise: Deploying New Versions with Zero Downtime
- Automation
- CI/CD Pipelines
- Source and Version Control
- Automated Testing
- Automated Builds
- Deployment
- Application Monitoring and Alerting
- Monitoring
- Logging
- Tracing
- Uptime Checks
- Alerts
- Exercise: Uptime Checks
- IAM
- User Accounts
- Service Accounts
- Roles and Permissions
- Exercise: Controlling App Permissions with IAM
- Network Security
- Securing Cloud Networks
- Firewalls and Security Groups
- IP Addresses
- Mitigating DDoS Attacks in the Cloud
- Exercise: Configuring Networks and Firewalls
- Data Security
- Encryption
- Key Management Services in the Cloud
- Secret Managers
- Data Loss Prevention
- Exercise: Using Secret Managers
- Big Data Processing
- Overcoming Limitations with On-Premises Systems
- Distributed Data Processing
- MapReduce
- Hadoop and Spark
- Data Analytic Services
- Machine Learning
- Understanding Cloud-Based Machine Learning Services
- Leveraging Pretrained Machine Learning APIs
- Exercise: Data Loss Prevention with ML
- Planning Application Migrations
- Application Inventory
- Dealing with Dependencies
- Application Prerequisites
- Prioritization
- Choosing First Movers
- Activity: Planning Application Migration
- Migration Strategies
- Lift and Shift
- Move and Improve
- Refactor
- Rewrite
- Activity: Selecting a Migration Strategy
- Data Migration Strategies
- Analyzing Downtime
- Moving the Data
- Backup and Restore
- Incremental Backup and Restore
- Continuous Replication
- Split Read/Writing
- Data Access Microservice
- Activity: Planning Database Migrations
- The Importance of Automation
- Automating Migration Activities
- Infrastructure as Code
- Capacity Planning
- Resource Monitoring
- Cost Estimation
- Activity: Estimating the Cost of Migration
