LLMs Unleashed -
An IT/DevOps Adventure with AWS Bedrock

09:30-10:15

Gathering & Breakfast

10:15-10:45

Introduction to LLM and Prompt Engineering
Understand the fundamentals of Large Language Models and learn various prompt engineering patterns to optimize your use of LLMs

10:45-11:30

Explore AWS LLM Offerings
Discover Amazon Bedrock and SageMaker Jumpstart, including deep dive into vector embedding and RAG

11:30-11:45

Coffee Break

11:45-13:15

Hands-On Labs (1.5 hours)

• Prompt Engineering Lab: Apply learned patterns to optimize your LLM usage
  
  
• Entity Extraction Lab: Utilizing entity extraction patterns for parsing heterogeneous log files to extract critical data like IP addresses, instance IDs, service names, error codes, and configuration details.
  
  
• Knowledge Bases and Retrieval-Augmented Generation (RAG) Lab: Enhancing AI capabilities through knowledge bases and RAG techniques. This involves integrating AWS Well-Architected Framework documentation into a knowledge base and analyzing IAM policies for best practices.

13:15-13:45

Lunch

13:45-15:15

Hands-On Labs (1.5 hours)

• Converting Terraform Templates to AWS with Bedrock: Streamline infrastructure deployment by converting Azure Terraform templates to AWS.
  
  
• Manual Creation and Testing of a Guardrail Using the AWS Console: Walk through the manual creation and testing of a guardrail to ensure robust and secure deployments.
  
  
• Monitoring Generative AI Applications: Integrate Amazon Bedrock with Amazon CloudWatch for effective monitoring.

15:15-15:30

Coffee Break

15:30-16:00

Summary and Future Directions
Discuss key differences between AWS LLM offerings and explore exciting future advancements in production environments.