A journey through the AWS ecosystem that democratizes access to Generative AI

AWS offers many innovative services that lower the barriers to entry for starting to experiment with generative AI. In this session, we will explore the use of PartyRock to create prompt-based web apps; we'll discover how to leverage Amazon Bedrock to invoke generative AI models via API, for example, creating an app to generate Infrastructure as Code from architectural drawings; finally, we'll introduce the brand new Bedrock Studio, which allows you to create apps in minutes with advanced agent and guardrails features. Thanks to these tools, AWS democratizes access to generative AI in less than an hour, regardless of the initial skill level.

Transcript

1. A journey through the AWS ecosystem that democratizes access to Generative AI Monica Colangelo COM306 (she/her) Principal Cloud Architect NTT DATA
3. Initial training Fine Tuning (PEFT, QLoRA) Full Fine Tuning Advanced RAG Basic Retrieval Augmented Generation (RAG) Prompt Engineering (single/few/advanced) Use LLM as is (Zero Shot) Users AI scientists MLOps Engineers ML Engineers Principal Data Scientists Data Engineers Data Stewards Data Governance Data Scientists App Developers Junior Data Scientists Prompt Engineers Higher performance Higher Cost $$ Higher Complexity Specific Skills Easy Implementation Low Adaptability The thicker, the more used
4. PARTYROCK
5. AMAZON BEDROCK 01 04 02 05 03 06 Access a range of leading FMs through a single API Privately customize FMs with your own data Extend the power of LLMs with your data using RAG Build agents that execute complex business tasks by dynamically invoking APIs Enable data security and compliance Get the best price performance without managing infrastructure
6. The easiest way to build and scale generative AI applications with foundation models Summarization, complex reasoning, writing, coding Contextual answers, summarization, paraphrasing High-quality images and art Text generation, search, classification Q&A and reading comprehension Text summarization, generation, Q&A, search, image generation Amazon Titan Text Premier Amazon Titan Text Lite Amazon Titan Text Express Amazon Titan Text Embeddings Amazon Titan Text Embeddings V2 Amazon Titan Multimodal Embeddings Amazon Titan Image Generator Claude 3 Opus Claude 3 Sonnet Claude 3 Haiku Claude 2.1 Claude 2 Claude Instant Llama 3 8B Llama 3 70B Llama 2 13B Llama 2 70B Command Command Light Embed English Embed Multilingual Command R+ Command R Stable Diffusion XL1.0 Stable Diffusion XL 0.8 Jurassic-2 Ultra Jurassic-2 Mid Mistral Large Mistral 7B Mixtral 8x7B Text summarization, Q&A, text classification, text completion, code generation AMAZON BEDROCK
7. BEDROCK STUDIO
8. BEDROCK STUDIO
9. BEDROCK STUDIO
10. BEDROCK STUDIO
11. CUSTOM LOCAL APP USING BEDROCK
12. CUSTOM LOCAL APP USING BEDROCK
13. CUSTOM LOCAL APP USING BEDROCK
14. A REAL SOLUTION - UNDER THE HOOD
15. A REAL SOLUTION - UNDER THE HOOD
16. A REAL SOLUTION - UNDER THE HOOD
17. A REAL SOLUTION - UNDER THE HOOD
18. A REAL SOLUTION - UNDER THE HOOD
19. NOW GO BUILD CODECATALYST GITHUB LINKEDIN AWS COMMUNITY DAY ITALY GenAI chatbot blueprint Basic PoC code Let’s connect! September 27th , Roma BLOG Bedrock Studio First Look Transform diagrams in IaC ...and more www.awscommunityday.it