Build On Live | Generative AI Special - May 16, 2024

Hello friends and welcome to the Build On Live | Generative AI Special show notes!

TODAY, We’re exploring 3 key concepts throughout the day: 1/ Generative AI Foundations 1; 2/ LLM Fine-Tuning & Customization; and 3/ Using Generative AI to Build End-to-End Applications.

Further your learning around each concept by exploring the curated resource collection, comprising blog posts (📝), video tutorials (🎥), and AWS Skill Builder courses (📓):

Hop on to 🐰:

In the first part of the day, we dove into the foundations of Generative AI, navigating through the AWS Gen AI Stack. We covered selecting the right model for your use case by price, performance, and capability using Amazon Bedrock. Additionally, part of the foundation includes best practices in prompt engineering for LLMs, and strategies to save you time using Generative AI tools like Amazon Q.

Generative AI Primers:
📝 Can generative AI only improve from here? It's complicated
🎥 Deep Dive into LLMs - with AWS!
📓 Introduction to Generative AI: Art of the Possible
📓 Planning a Generative AI Project
📓 Learn Generative AI - for Developers!

Prompt Engineering How-Tos:
📝 Master Prompt Engineering: Build AI apps with Claude & PartyRock
📝 How to Prompt Mistral AI models, and Why
📝 Handling Time Management: I built a scheduling assistant with Claude and Streamlit
📓 Foundations of Prompt Engineering

Get to Know the AWS Generative AI Stack:
🎥 Getting Started with Amazon Bedrock
📓 Amazon Bedrock Getting Started
📓 Amazon Q Introduction

By the end of this segment, attendees will have refined their skills in fine-tuning and evaluating Large Language Models (LLMs) using Amazon Bedrock. This will enable them to customize foundational models (FMs) securely and privately with their own data. Moreover, participants will become proficient in using Retrieval-Augmented Generation (RAG) to boost AI search capabilities with vector datastores. They'll also learn to apply RAG to practical scenarios like drug discovery by leveraging extensive knowledge bases integrated with Amazon Bedrock.

Applying RAG in Real-Life:
📝 I built an AI landmark recommender for visiting new cities
📝 GenerativeAI & AWS: Building Chatbot with a Knowledge Base
📝 Dolly, Miley, or Beyonce? I asked the new Chat With Your Document feature in Amazon Bedrock.
📝 How to use Retrieval Augmented Generation (RAG) for Go applications
🎥 Improve your Generative AI Application with RAG
📓 Introduction to Information Retrieval with Amazon Q

More Fine-tuning & Customization:
📝 What is an instruct model? - Instruction and Chat Fine-Tuning

In this module, participants will gain the expertise to construct comprehensive applications powered by Generative AI. They will also learn how to simplify the development process using agents on Amazon Bedrock, facilitating more efficient creation, deployment, and management of AI-driven applications.

Amazon Bedrock Agents:
📝 Building an AWS Solution Architect Agent with Amazon Bedrock
🎥 Agents Tools & Function Calling with Amazon Bedrock (How-to)
🎥 Build an AWS Solutions Architect Agent with Amazon Bedrock

Automating Games Using Chain of Thought:
📝 14 LLMs fought 314 Street Fighter matches. Here's who won.
📝 Super Mario Bros: The LLM Levels - Generate levels with a prompt
🎥 Let's Ship It - with AWS! Generative AI