Build and Deploy GenAI RAG Applications and AI Agents using Google Cloud

​We are super excited to collaborate with the Google team to bring you the latest on AI agentic application development! It’s a great opportunity to learn and connect with your fellow developers in the Bay Area. Food and drinks will be provided.

​Agenda
5:00 - Open doors & mingling
5:45 - Welcome and brief intros
6:00 - Talks
7:30 - Socializing
8:00 - Close doors

​Talk #1 Beyond Vectors with Mark Quinsland
Retrieval Augmented Generation (RAG) is an essential tool for limiting the hallucinations of LLMs. Unfortunately, using a vector database for RAG introduces its own limitations. The Gartner consulting organization has recommended using Knowledge Graphs for RAG when accurate and safe result generation is required. GraphRAG is a solution that combines the results of a Knowledge Graph search with a vector search. The vector typically is used to find similar documents and the Knowledge graph then helps improve the scoring that determines which of the candidate documents should be returned. This left-brain:right-brain combination produces answers that are not only more accurate but that can also be curated for the user.
A Colab notebook will be shown to compare the results of queries using a vector-only RAG search with a GraphRAG search. The demo will show the same query being answered by a GraphRAG application with Neo4j on the Vertex AI reasoning engine.

​Talk #2 Introduction to Reasoning Engine with Vertex AI with Ran Li
The Reasoning Engine provides a managed runtime for your customized agentic workflows in generative AI applications. You can create an application using orchestration frameworks such as LangChain, and deploy it with Reasoning Engine.
In this session, Ran will demonstrate how to use Reasoning Engine with open-source frameworks and discuss how to develop, deploy, query, and manage AI agents using Vertex AI.

​Talk #3 Building AI Agents with Maruti C
Generative AI and AI agents are revolutionizing application workflows. Increasingly, engineering teams deploy intelligent agents to reduce process inefficiencies, human errors, and manual tasks.
An AI platform like Vertex AI can help you to easily build and deploy these agents. This session will explore various options to build, deploy, and monitor AI agents on Vertex AI, with a special focus on Agentic RAG and GraphRAG on Google Cloud.