AI Agents with Elastic Agent Builder & Introduction to Elastic Inference Service

Join us for a meetup at Motorworld on Wednesday November 5th, the eve before ElasticON!

We'll have presentations, networking, pizza and refreshments.

Address
Motorworld Munich - Stellwerk 2
Am Ausbesserungswerk 8 · München, BY
When you arrive, please inform your name and that you will go to the meetup at Stellwerk 2.

Agenda
18.00: Doors Open, join us for pizza and a drink

18.20: Talk #1: Building AI Agents with Elastic Agent Builder and MCP
Tatiana Finamor, Customer Architect at Elastic
Abstract: The Model Context Protocol (MCP) creates a bridge between AI models and external tools, allowing them to exchange information in a structured and secure way.
In this session, we will look at how Elastic fits into this new ecosystem.
We will demonstrate how to use Claude Desktop to query Elastic data through MCP using natural language, and how the Elastic Agent Builder can be used within a serverless project to build and deploy intelligent, data-aware agents.
This session is a practical introduction for anyone curious about combining Elastic with modern AI frameworks to simplify data access, automation, and intelligent interaction.

19.00: Talk #2: Intro to the Elastic Inference Service (EIS)
Max Jakob, Principal Software Engineer I - Search Experiences at Elastic
Tim Grein, Senior Software Engineer at Elastic
Abstract: AI models are everywhere in search these days. In this talk, we will present the Elastic Inference Service (EIS), the new way to apply AI models within the Elastic Stack. This GPU-backed service enables high throughput ingest, low latency search, reranking and LLM chat that all integrate neatly with Elasticsearch and Kibana. Everything is managed by Elastic, no need for users to configure any infrastructure or models. Get started with semantic search and built-in Kibana AI features out-of-the-box! We will demo what's possible today and give an outlook into what's coming in the next months.

19.40 Lightning Talk Evolution of search - from keyword to contextual
Petar Perković, Data Science and DWH Engineering Manager at Njuškalo
Nikolina Kosanović, Senior ML Engineer at Njuškalo
Abstract: Search technology has come a long way from simple keyword matching to systems that truly understand meaning and context. In this talk, we’ll trace that evolution - from traditional keyword search to semantic and contextual search, and explore how each approach impacts relevance and user experience. Using real-world examples and practical comparisons, we’ll uncover how these techniques work under the hood, where Elastic fits in and how we use it at Njuškalo.

19.55: Networking

20.30: Close

PS: Join ElasticON Munich the next day — a free, full-day conference on search, observability, and security with Elastic, you can register here.