AWS AI Meetup: Generative AI, LLMs and Vector DB

Pre-registration is required for admission: https://www.aicamp.ai/event/eventdetails/W2023092615

(Venue security may not let you in if you don't pre-register at the link above)

Welcome to our in-person ML monthly meetup in Chicago, in collaboration with AWS. Join us for deep dive tech talks on AI/ML, food/drink, networking with speakers & peers developers.

Agenda:

* 5:00pm~5:30pm: Checkin, Food/drink and networking

* 5:30pm~5:40pm: Welcome/community update/Sponsor intro

* 5:40pm~7:30pm: Tech talks

* 7:30pm: Open discussion & Mixer

Tech Talk 1: Choice of Vector Database for GenAI

Speaker: Puneet Rawal @AWS

Abstract: In this session Puneet Rawal will walk through various factors that goes into picking a vector database for GenAI use cases like ChatBot, Summarization, Content Generation etc. This will be an in-depth look at comparison of index types, configuration, performance and cost of various databases like Postgres, OpenSearch, PineCone, Milvus etc.

Tech Talk 2: Building a GenAI-based chat bot using Retrieval Augmented Generation

Speakers: Venu Thangalapally, Guillermo Tantachuco @AWS

Abstract: In this session, we will describe GenAI, Retrieval Augmented Generation (RAG), and how RAG ensures that responses are grounded in factual information to reduce the likelihood of hallucinations. We will use Amazon Bedrock to build a question answering chat bot with foundation models (FMs). This chat bot leverages RAG, AWS Lambda, and Amazon Kendra to search a knowledge base and understand the context and return the most relevant results.

Tech Talk 3: Fine Tuning LLMs to augment IDP workflows

Speakers: Alfredo Castillo, Arun Batra @AWS

Abstract: Performing Common Sense Reasoning and question answering using Large Language Models (LLMs) on a document is a crucial task in Natural Language Processing (NLP) to extract meaningful context and provide insightful answers from the textual data in a more intuitive manner. In this lecture, we will explore how to extract text from documents using Amazon Textract for Intelligent Document Processing workflows (IDP). Improve the responses via Retrieval-Augmented Generation (RAG) approach with Vector database and create inferences from the LLM for Q&A.