SF Unstructured Data Meetup May 21 2024

🎥 Once a month, we'll meet, socialize, and hear speakers present topics on unstructured data and generative AI. This event was sponsored by Zilliz.

Timeline:
01:17 - A welcome from Charles Xie, Zilliz CEO & Founder
05:48 - Speaker Christy Bergman, Introduction to Open Source RAG and RAG Evaluation
29:51 - Speaker Charles Packer, Introduction to Memory Augmented Chat
59:26 - Speakers Idan Gazit and Cole Bemis, Copilot Workspace: What it is, how it works, why it matters

~~~~~~~~~~~~~~~ CONNECT ~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~ MEETUP VIDEO CONTENTS ~~~~~~~~~~~~~~

1. Host & Speaker: Christy Bergman, Developer Advocate at Zilliz
Title: Introduction to Open Source RAG and RAG Evaluation
Abstract: You’ve heard good data matters in Machine Learning, but does it matter for Generative AI applications? Corporate data often differs significantly from the general Internet data used to train most foundation models. Join me for a demo on building an open source RAG (Retrieval Augmented Generation) stack using Milvus vector database for Retrieval, LangChain, Llama 3 with Ollama, Ragas RAG Eval, and optional Zilliz cloud, OpenAI.

2. Speaker: Charles Packer, Co-founder, MemGPT
Title: Introduction to Memory Augmented Chat

3. Speakers: Idan Gazit and Cole Bemis
Title: Copilot Workspace: What it is, how it works, why it matters
Abstract: Copilot Workspace recently launched into technical preview! Bring your popcorn and come see a live demo of it in action. We'll talk a little bit about the AI developer tooling landscape, what comes after chat, and how these tools should help us to create.

Title: Build an agent using prompts and Notion-like commands
Demo was how to create 4 different agents. (1) Q&A about a legal contract with Zilliz pipelines, (2) enhance Q&A with advanced RAG features, (3) use a Chain-of-Thought agent to find Bitcoin prices, and (4) conduct comprehensive internet research to create detailed content, summary, poem, image, and website.