Inside Tinder’s real-time recommendation system to see how they use Elasticsearch at a massive scale:

What did you build with Elasticsearch in 2025? Share your creation below!

Do you know how to configure synonyms in Elasticsearch? There are a few ways, like listing them in a file, but the Synonyms API is an easy way to centrally manage them by specifying your set and updating them when needed (1 & 2). You can then use it by specifying the set in your index analyzer (3).

Ever wondered how Tinder finds you? Check out how Tinder uses geo-indices to power mass-scale matching:

Elastic currently provides inference endpoints for several well-known third-party providers, like OpenAI & Azure. But did you know you can also define your own custom endpoints using the inference API? Check out this example to generate embeddings using DeepSeek. Learn more: go.es.io/45c4TNQ

When demonstrating in Kibana Dev Tools, we often work with API keys. To avoid exposing them, we can use variables to define reusable placeholders for dynamic values in queries. These variables, persisted via browser local storage, can then be referenced directly in the commands we run in Dev Tools.

Spotify Wrapped is a data analytics story (with a great PR team). Fellow community members, we know a lot of people were unhappy with how theirs came out… So don’t worry, we’ve got you. We’ll show you how to build your own Spotify Wrapped with Kibana. Try it now: www.elastic.co/search-labs/...

How to make your own Spotify Wrapped in Kibana - Elasticsearch Labs

Here's how you can make your own Spofity Wrapped in Kibana with the top artists, songs, and trends over the year.

elastic.co

Hybrid search🕵️‍♂️ brings together the precision of lexical search and the intuition of semantic search. It uses both traditional scoring and vector similarity to surface the right results at the right time and help you build search that actually understands your data.

New on Elasticsearch Labs: the impact of relevance on context engineering. Discover how hybrid retrieval, semantic chunking, and agentic search improve LLM performance by managing context relevance efficiently ⬇️ www.elastic.co/search-labs/...

Context engineering VS prompt engineering: what's the difference? Think of prompt engineering as writing the instruction, and context engineering as building the full environment so the AI can follow it.

Go deeper into the architecture and reasoning behind Elastic Agent Builder — how we integrated vector search, retrieval relevance, and open standards to create a foundation for building data-driven AI agents. Learn more in the post: elastic.co/search-labs/...

Expose your Elastic agent and tools as an MCP server to integrate with your preferred client with Elastic Agent Builder, or build your own interface.

Justin Castilla@justincastilla.bsky.social · 11mo ago

@elastic.co 's Agent Builder is basically a Build-A-Bear store for creating custom #a2a agents and #MCP server tools and I love it! I have 100k of manufacturing logs and it chomped through them like it was nothing. I love this evolution of LLM-focused data retrieval! ela.st/agent-builde...

Elastic Agent Builder demo 🙌

🥼 Piotr Przybył@piotrprz.bsky.social · 11mo ago

The new Agent Builder from @elastic.co helps not only data analysts, but also developers. Don't know which data to fetch from #Elasticsearch and how? Worry not, just use a coding agent to prototype working solution.

The new Agent Builder from @elastic.co helps not only data analysts, but also developers. Don't know which data to fetch from #Elasticsearch and how? Worry not, just use a coding agent to prototype working solution.

Build a custom tool for your agents with Elastic Agent Builder. Next demo: Agent Builder and MCP

All you need is a natural language agent definition, and you have a custom AI assistant to help you talk to your Elastic data thanks to Agent Builder.

Loving these demos of everything you can build with Elastic Agent Builder 🚀

Iulia Feroli@iuliaferoli.com · 12mo ago

After announcing the @elastic.co Agent Builder on our YouTube and dev (@elastic-dev.bsky.social) channels, I played around with it some more to build a #SpotifyWrapped generator based on a previous blog I wrote! I build my own ES|QL tool and musical agent 🎵

Elastic Observability is using unsupervised ML + GenAI assistants to turn logs into real-time incident response intelligence. Walk through a step‑by‑step investigation that detects anomalies, surfaces root causes, measures user impact, and accelerates remediation. www.elastic.co/observabilit...

Follow along to create a custom agent with Elastic Agent Builder. Next demo: How to create custom tools

Agent Builder demo by @carlyrichmond.bsky.social 🚀

Carly Richmond@carlyrichmond.bsky.social · 12mo ago

Have you played with the new @elastic.co Agent Builder yet? I've been fortunate enough to have an early play with it to solve the problem of finding the set that elusive LEGO piece belongs to 🧱. Check out how it works in the brief video 👇. #ai #elasticsearch #contextengineering

Want to chat with your data? Pick an index and walk through this step-by-step notebook to create a tool and build an agent in 10 minutes with Elastic Agent Builder: github.com/elastic/elas...

Introducing Elastic Agent Builder, designed to give your agents the best context and tools. Build precise agents fast, connect all your data where it already lives in Elasticsearch, and create custom tools for any use case. Learn more: www.elastic.co/search-labs/...

👋 Welcome to Elastic Dev — your new go-to for hands-on demos, cheat sheets, concept explainers & more. Get the latest on search, data & GenAI straight from the source. Expect: - Practical demos & how-tos - Posts & technical deep dives - Open source updates + more Ready, set, build. Hello, world.