Connecting external data sources to LLMs to give them memory is a crucial part of many LLM applications. This can take the form of connecting to pre-existing large corpora of data, remembering user conversations, or even creating “new memories” by reflection. Underpinning all of all this is retrieval - the process of pulling relevant pieces of information into context.

Join LangChain’s Harrison Chase for a deep dive into retrieval where we dissect the challenges of finding relevant information from a large corpus of data. 

You’ll learn:
- What is memory and why is it important
- Types of memory
- Basics of semantic search
- Edge cases of semantic search
- Generative Agent examples

# About the session
Connecting external data sources to LLMs to give them memory is a crucial part of many LLM applications. This can take the form of connecting to pre-existing large corpora of data, remembering user conversations, or even creating “new memories” by reflection. Underpinning all of all this is retrieval - the process of pulling relevant pieces of information into context.

Join LangChain’s Harrison Chase for a deep dive into retrieval where we dissect the challenges of finding relevant information from a large corpus of data. 

## You'll learn
- What is memory and why is it important
- Types of memory
- Basics of semantic search
- Edge cases of semantic search
- Generative Agent examples

meetup.svg

Webinar

Join LangChain's Harrison Chase for a live session exploring different retrieval techniques for getting the most relevant context from your LLM application.

Memory for LLM Applications with LangChain's Harrison Chase | Zilliz Webinar

# What to Except
You’ll receive an email from Zoom with the details to login for the session. We look forward to seeing you at the upcoming webinar!

- Thursday, June 8, 2023
- 9:00 AM Pacific | 12:00PM Eastern
- 45-minute presentation followed by audience questions

Harrison Chase

Memory for LLM applications: Different retrieval techniques for getting the most relevant context

In this hands-on tutorial, we’ll introduce LLMs and two main problems they face when it comes to production. First, high cost. Second, lack of domain knowledge. We then introduce vector databases as a solution to this problem. We cover how a vector database can facilitate data injection and caching through the use of vector embeddings.

Then we’ll use this knowledge to build an LLM application using LlamaIndex and Milvus, the world’s most popular vector database.

What you’ll need:
- Python 3.9 or above
- A basic understanding of vectors and databases

What you’ll learn:
- What is a vector database
- Why do LLMs face data issues?
- How to deal with data issues in an LLM

# What will you learn
In this hands-on tutorial, we’ll introduce LLMs and two main problems they face when it comes to production. First, high cost. Second, lack of domain knowledge. We then introduce vector databases as a solution to this problem. We cover how a vector database can facilitate data injection and caching through the use of vector embeddings.

Then we’ll use this knowledge to build an LLM application using LlamaIndex and Milvus, the world’s most popular vector database.

What you’ll need:
- Python 3.9 or above
- A basic understanding of vectors and databases

## Topics covered
- What is a vector database
- Why do LLMs face data issues?
- How to deal with data issues in an LLM

Training

Join us for a hands-on tutorial building LLM applications that scale using Milvus and LlamaIndex.

Tutorial: Working with LLMs at Scale | Zilliz Webinar

# What to Except
You’ll receive an email from Zoom with the details to login for the session. We look forward to seeing you at the upcoming webinar!

- Thursday, June 15, 2023
- 9:00 AM Pacific | 12:00 PM Eastern
- Hands-on tutorial

## What you'll need
- Python 3.9 or above
- Code editor like VS Code
- A basic understanding of vectors and databases

Yujian Tang

Tutorial: Working with LLMs at Scale

Join us at the LLMs in Production virtual conference by MLOps on the 15th-16th of June. Our very own Yujian Tang will be speaking about "Solving LLM Data Problems" on Friday, June 16th at 12pm Pacific. Register free to join us! 

live.svg

Virtual Event

LLMs in Production - Part II

In this hands-on tutorial, we’ll introduce embeddings and vector search from both an ML- and application-level perspective. We’ll start with a high-level overview of embeddings and discuss best practices around embedding generation and usage. Then we’ll use this knowledge to build a semantic text search application. Finally, we’ll see how we can put our application into production using Milvus, the world’s most popular open-source vector database.

What you’ll need:
- Python 3.9 or above
- A basic understanding of vectors and databases

What you’ll learn:
- What is vector database
- What is semantic similarity
- How to use a vector database to find similar texts

# What will you learn?
In this hands-on tutorial, we’ll introduce embeddings and vector search from both an ML- and application-level perspective. We’ll start with a high-level overview of embeddings and discuss best practices around embedding generation and usage. 

Then we’ll use this knowledge to build a semantic text search application. Finally, we’ll see how we can put our application into production using Milvus, the world’s most popular open-source vector database.

## What you’ll need:
- Python 3.9 or above
- A basic understanding of vectors and databases

## What you’ll learn:
- What is vector database
- What is semantic similarity
- How to use a vector database to find similar texts


Get hands-on in this live tutorial where you'll learn how to use a vector database to build a semantic text search application.

Tutorial: Building a Semantic Text Search Application | Zilliz Webinar

Tutorial: Building a Semantic Text Search Application

ChatGPT, the artificial intelligence chatbot from OpenAI, has taken the world by storm and has transformed the way we perform everyday tasks. With its advanced language processing capabilities, it has become a go-to solution for people looking for quick answers, whether it's about the weather, the news, or how to cook a recipe. 

New open source plugins have made it possible for developers to easily build their own ChatGPT knowledge base solutions. At the heart of these solutions is a data model that converts unstructured data from text, videos, images and more into vector embeddings. 

Vector databases are designed to efficiently store and retrieve these high-dimensional embeddings, making them an ideal solution for chat applications that require fast and accurate vector search results. Join Frank Liu, ML architect at Zilliz, for a session on why vector databases are critical to the success of these LLM-based chat solutions and how developers can build powerful chat solutions that deliver quick answers.

In this session, you'll learn:

- What is a vector database
- Why is it important to store your embeddings in a purpose-built database
- Why dumping your embeddings into Postgres is a bad idea
- How we built a chat knowledge base for open-source projects using Zilliz, prompts-as-code, and ChatGPT

The Rising Demand for Vector Databases in the Age of ChatGPT

The Hierarchical Navigable Small World (HNSW) is one of the most popular graph-based indices for nearest neighbor search. Join us for a fireside chat with Yury Malkov, as we discuss how he ended up developing this critical method of high-dimensional data analysis. We’ll explore how he got started in computer science and where he sees the future of vector search and LLMs. You’ll even get a chance to ask Yury questions at the end of the chat, you won’t want to miss it!

Topics covered:
- Why HNSW and other graph indexes are great tools for vector search
- What is the future of vector search?
- Thoughts on the proliferation and increasing importance of LLMs as it relates to vector data

## About the Session
The Hierarchical Navigable Small World (HNSW) is one of the most popular graph-based indices for nearest neighbor search. Join us for a fireside chat with [Yury Malkov](https://twitter.com/malkovyury), as we discuss how he ended up developing this critical method of high-dimensional data analysis. We’ll explore how he got started in computer science and where he sees the future of vector search and LLMs. You’ll even get a chance to ask Yury questions at the end of the chat, you won’t want to miss it!

# Topics covered:
- Why HNSW and other graph indexes are great tools for vector search
- What is the future of vector search?
- Thoughts on the proliferation and increasing importance of LLMs as it relates to vector data

Register for a free webinar with Yury Malkov, author of HNSW and hear his predictions on the future of vector search.

Fireside chat with HNSW author Yury Malkov | Zilliz Webinar

Yury Malkov

What is the future of vector search? A fireside chat with HNSW author Yury Malkov

Join Frank Liu at Open Source Summit North America for The Rise of Vector Databases - Lessons From the Milvus Community.

The total amount of digital data generated worldwide is increasing at a rapid rate. Simultaneously, approximately 80% (and growing) of this newly generated data is unstructured data - data that does not conform to a table- or object-based model. Examples of unstructured data include text, images, protein structures, geospatial information, and IoT data streams. Despite this, the vast majority of companies and organizations do not have a way of storing and analyzing these increasingly large quantities of unstructured data. Embeddings - high-dimensional, dense vectors which represent the semantic content of unstructured data - can remedy this. Armed with this knowledge, it's clear that the mobile/IoT era necessitates a new type of cloud-native, fully distributed database purpose-built to store, search, and index large quantities of embedding vectors: the vector database.

In this presentation, we'll introduce some of the interesting production use cases we've seen with Milvus - the world's most popular open source vector database. Additionally, we'll discuss pitfalls and challenges associated with integrating Milvus into a production data/ML stack. Finally, we'll wrap up with a discussion on ongoing community feedback and future plans for Milvus.

The Rise of Vector Databases - Lessons From the Milvus Community

In-context learning has emerged as a powerful method to augment LLMs with our own data. By inserting relevant context into the input prompt, we can leverage the LLM's sophisticated reasoning abilities to generate accurate and meaningful responses that reflect our unique data inputs. 

LlamaIndex is a simple, flexible interface to connect your LLM's with external data. Join this session with LlamaIndex co-founder and CEO, Jerry Liu. He’ll share details on how he built LlamaIndex, discuss the various tools it contains and how you can get started enriching your LLM with private data.

You'll learn:
- How LlamaIndex can help augment your LLM with private data
- What LlamaIndex tools exist for data connection, indices and queries 
- Best practices for integrating LlamaIndex with Milvus

# About the Session
In-context learning has emerged as a powerful method to augment LLMs with our own data. By inserting relevant context into the input prompt, we can leverage the LLM's sophisticated reasoning abilities to generate accurate and meaningful responses that reflect our unique data inputs. 

LlamaIndex is a simple, flexible interface to connect your LLM's with external data. Join this session with LlamaIndex co-founder and CEO, [Jerry Liu](https://twitter.com/jerryjliu0). He’ll share details on how he built LlamaIndex, discuss the various tools it contains and how you can get started enriching your LLM with private data.

## What you'll learn:
- How LlamaIndex can help augment your LLM with private data
- What LlamaIndex tools exist for data connection, indices and queries 
- Best practices for integrating LlamaIndex with Milvus

Zilliz webinar covering how to boost your LLM with private data with LlamaIndex to generate accurate and meaningful responses that reflect unique data inputs.

Boost your LLM with Private Data Using LlamaIndex | Zilliz Webinar

Jerry Liu

Boost your LLM with Private Data using LlamaIndex

It's no secret that ChatGPT, the artificial intelligence chatbot from OpenAI, has taken the world by storm and is reinventing how most of us complete everyday tasks. Now that OpenAI has open-sourced retrieval plugins for ChatGPT, allowing anyone to develop their own knowledge base retrieval app, expect to see an explosion of solutions helping people access information at lightning speed.

At the core of these chat solutions, data models transform unstructured data into vector embeddings stored in a vector database. With these embeddings, developers can perform similarity searches to provide the most relevant answers to their users. Join Frank Liu, ML architect at Zilliz, for a session on why vector databases are critical to the success of these LLM-based chat solutions.

How ChatGPT Is Fueling the Need for Specialized Vector Storage

Learn how to improve ChatGPT apps by addressing two common issues: incorrect answers due to lack of knowledge, and slow response times due to high query volume. Discover how using a vector database can help you solve both issues allowing you to provide domain-specific knowledge to your LLM application and cache information to speed up follow-up queries. Learn how vector databases can store, query, and retrieve embeddings for similarity searches quickly and at scale. We’ll share free tools to help you build your application for optimal performance and scale immediately! 

Got embeddings? Discover the key to building AI applications that scale

Zilliz is excited to announce the general availability of Zilliz Cloud, a fully-managed Milvus service made by the creators of Milvus. This new multi-cloud service (AWS and GCP) offers consistent management, security, and governance experience across all clouds. It provides reliability & resiliency built-in, 99.99% uptime SLA, and zero data corruption. Frank Liu, Director of Operations & ML Architect at Zilliz, will demonstrate key product features in this webinar. 

He will dive into:
- New cluster sizes to support over a billion multi-dimensional vectors
- Embedding model integrations (OpenAI, Hugging Face, Cohere)
- Powerful Backup and restore
- Simple Migration from Milvus to Zilliz
- All the latest Milvus features
- Performance improvements and more!

# What will you learn?
Zilliz is excited to announce the general availability of Zilliz Cloud, a fully-managed Milvus service made by the creators of Milvus. This new multi-cloud service (AWS and GCP) offers consistent management, security, and governance experience across all clouds. It provides reliability & resiliency built-in, 99.99% uptime SLA, and zero data corruption. Frank Liu, Director of Operations & ML Architect at Zilliz, will demonstrate key product features in this webinar. 

## Topics covered:
- New cluster sizes to support over a billion multi-dimensional vectors
- Embedding model integrations (OpenAI, Hugging Face, Cohere)
- Powerful Backup and restore
- Simple Migration from Milvus to Zilliz
- All the latest Milvus features
- Performance improvements and more!

Learn the differences between Milvus and Zilliz Cloud and understand how to start building vector search applications.

Unlocking the power of vector search with Zilliz Cloud | Zilliz Webinar

Frank Liu

Unlocking the power of vector search in Zilliz Cloud

This session is being presented as a part of the [Future of Data Science & AI](https://hubs.la/Q01D9l2W0) virtual conference hosted by Data Science Dojo on March 1-2, 2023.

The total amount of digital data generated worldwide is increasing at a rapid rate. Simultaneously, approximately 80% (and growing) of this newly generated data is unstructured data – data that does not conform to a table- or object-based model. Examples of unstructured data include text, images, protein structures, geospatial information, and IoT data streams. Despite this, the vast majority of companies and organizations do not have a way of storing and analyzing these increasingly large quantities of unstructured data. Embeddings – high-dimensional, dense vectors which represent the semantic content of unstructured data – can remedy this. 

In this tutorial, we’ll introduce embeddings and vector search from both an ML- and application-level perspective. We’ll start with a high-level overview of embeddings and discuss best practices around embedding generation and usage. We’ll then use this knowledge to build two systems: semantic text search and reverse image search. Finally, we’ll see how we can put our application into production using Milvus, the world’s most popular open-source vector database.



Unlock the Power of Embeddings with Vector Search

Learn about Milvus v2.0 and the groundbreaking functionality developed by the Milvus core team. v2.0 includes feedback from over 1000+ enterprise users and 10,000 community members in order to build a better database to serve enterprises in more AI and big data scenarios.

Meet Up

Enhancements Milvus v2.0

Topics for the meeting include: support for query/search return vector raw data, support for strings in Milvus 2.0, support for deletion in Milvus 2.0, and Milvus Insight and it's roadmap.


Live

Milvus Technical Meeting

The platform has already been adopted in local production by thousands of companies around the world, including Lucidworks, Shutterstock, and Cloudinary. In this webinar, Zilliz engineers will highlight the changes and features included with Milvus 2.0, a variety of use cases and applications in production—and how they empower small and large users alike.


Milvus 2.0, an open-source similarity search engine

Learn about Milvus v2.0 and the groundbreaking functionality developed by the Milvus core team. v2.0 includes feedback from over 1000+ enterprise users and 10,000 community members in order to build a better database to serve enterprises in more AI and big data scenarios.

After gathering this feedback and invaluable use cases, we are pleased to launch Milvus v2.0.

This version has made significant progress in Milvus as a cloud native and distributed system. In this talk, we will introduce the main features, functionality and future prospects of Milvus v2 .0



Introducing Milvus v2.0

Vector similarity search is powerful. It can extract valuable insights from complex information and make relevant recommendations from abstract inputs. But to be useful, it has to be fast. Whether suggesting products or surfacing valuable information, results often need to be served instantaneously.

How can you ensure similarity search is speedy, even when querying massive million, billion, or even trillion vector datasets? We've partnered with Lucidworks, an enterprise search technology company, to explain how the company used Milvus to build semantic search at speed. Join our webinar and discover how Milvus, an open-source vector database, makes it possible for Lucidworks to search an entire vector space in milliseconds.

Accelerating Semantic Search for AI-Powered Data Discovery with Milvus & Lucidworks

Searching big data is complex. Traditional keyword-based search methods are poorly suited for processing and analyzing unstructured data such as images, text, video, and audio that do not follow a predefined model. Vector similarity search utilizes deep learning methods and ANN (approximate nearest neighbor) algorithms to provide faster, more accurate search capable of processing trillion-scale unstructured datasets. This talk focuses on vector indexing, which plays a major role in accelerating time-consuming queries on large datasets. We will cover how various index types can be used in [Milvus](https://www.milvus.io/), a popular open-source vector similarity search engine, as well as Milvus v1.0 features and future plans for the platform.

Events

Events, conferences and webinars.

Upcoming Events